INDEX

Explanations

phrases related to communication or sending messages

oai_token-act-pair · gpt-3.5-turbo

the term "signal" in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

amily

-0.78

sm

-0.76

endor

-0.73

eatured

-0.72

frey

-0.71

hur

-0.70

eenth

-0.69

iler

-0.68

enne

-0.66

paragraph

-0.66

POSITIVE LOGITS

 signals

1.03

 signal

0.90

 signaling

0.90

 signalling

0.84

 Signal

0.83

 flares

0.82

 signs

0.79

 handlers

0.78

 signatures

0.74

 reinforcement

0.73

Activations Density 0.025%

phrases related to communication or sending messages

the term "signal" in various contexts

No Comments

No Known Activations