INDEX

Explanations

designations of specific entities or things

oai_token-act-pair · gpt-3.5-turbo

instances of the word "designated" and its variations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

Rh

-0.75

Bey

-0.73

ilant

-0.65

onse

-0.64

iga

-0.63

ENG

-0.62

pack

-0.61

Rate

-0.61

lb

-0.61

 decomp

-0.61

POSITIVE LOGITS

 designated

0.74

 zones

0.74

 markings

0.72

İĭ

0.71

 designate

0.69

arters

0.65

 assignments

0.64

 spots

0.64

eous

0.64

 Default

0.63

Activations Density 0.060%

designations of specific entities or things

instances of the word "designated" and its variations

No Comments

No Known Activations