INDEX

Explanations

words related to occupations or workplace environments

oai_token-act-pair · gpt-3.5-turbo

proper nouns and specific references to individuals or entities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

pling

-1.05

pled

-0.98

llan

-0.93

ãĥ£

-0.90

ples

-0.90

borgh

-0.81

celona

-0.80

hesda

-0.76

 Hour

-0.73

plings

-0.73

POSITIVE LOGITS

oid

0.88

age

0.85

 Strauss

0.83

ied

0.81

sylvania

0.79

hardt

0.79

iers

0.79

iest

0.78

etts

0.78

ational

0.78

Activations Density 0.071%

words related to occupations or workplace environments

proper nouns and specific references to individuals or entities

No Comments

No Known Activations