INDEX

Explanations

instances of the word "films"

oai_token-act-pair · gpt-3.5-turbo

references to films and movies

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

owder

-0.68

 legislature

-0.63

FUL

-0.60

 Fulton

-0.59

pse

-0.58

sie

-0.57

 Islanders

-0.57

GY

-0.57

ļé

-0.57

assis

-0.56

POSITIVE LOGITS

 movies

0.96

ensitive

0.95

uggest

0.95

earch

0.93

 films

0.93

chool

0.92

ilver

0.89

ynthesis

0.88

ettings

0.88

ovies

0.87

Activations Density 0.048%

instances of the word "films"

references to films and movies

No Comments

No Known Activations