INDEX

Explanations

sections of text indicating where to read more information

oai_token-act-pair · gpt-3.5-turbo

instances of interaction with media or prompts that lead to further content engagement

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.9.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 parity

-0.53

 differential

-0.52

iership

-0.46

 sailing

-0.44

 successive

-0.43

 sprawling

-0.43

 dividing

-0.42

fts

-0.42

 suites

-0.42

 fixtures

-0.41

POSITIVE LOGITS

.''.

0.86

.}

0.85

`.

0.77

 =================================================================

0.76

 ----------------------------------------------------------------

0.75

''.

0.75

________________________________________________________________

0.74

.).

0.72

].

0.71

 UNCLASSIFIED

0.71

Activations Density 1.958%

sections of text indicating where to read more information

instances of interaction with media or prompts that lead to further content engagement

No Comments

No Known Activations