INDEX

Explanations

proper nouns related to various entities like places, people, and organizations

oai_token-act-pair · gpt-3.5-turbo

mentions of specific locations or institutions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

 migrated

-0.57

 redistributed

-0.52

 adapted

-0.47

 renamed

-0.46

 reproduced

-0.46

tumblr

-0.45

uable

-0.45

 translated

-0.45

 added

-0.44

respective

-0.44

POSITIVE LOGITS

 Productions

0.50

gur

0.49

opa

0.47

 Therapy

0.46

 Strategies

0.46

 Holdings

0.45

eta

0.44

Corp

0.44

 Traps

0.44

Leaks

0.43

Activations Density 1.589%

proper nouns related to various entities like places, people, and organizations

mentions of specific locations or institutions

No Comments

No Known Activations