INDEX

Explanations

phrases related to authoritative statements or declarations

oai_token-act-pair · gpt-3.5-turbo

paragraph breaks in the text

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

ji

-0.56

azi

-0.55

elfth

-0.53

gh

-0.51

avia

-0.50

enium

-0.50

ene

-0.50

venth

-0.47

 sleeper

-0.46

rup

-0.46

POSITIVE LOGITS

————

1.08

————————

0.90

_-

0.84

please

0.72

again

0.61

feat

0.61

 Hide

0.61

)?

0.60

 namely

0.60

 thence

0.60

Activations Density 0.168%

phrases related to authoritative statements or declarations

paragraph breaks in the text

No Comments

No Known Activations