INDEX

Explanations

words related to specific tasks or objects within a larger context

oai_token-act-pair · gpt-3.5-turbo

specific nouns related to various contexts, including music, data, questions, and evidence

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

ctr

-0.60

 Recomm

-0.57

mbuds

-0.57

stood

-0.56

 Kills

-0.53

 Anniversary

-0.53

 advocacy

-0.51

 Tale

-0.51

 Canad

-0.51

 Nights

-0.51

POSITIVE LOGITS

 pertaining

0.96

 therein

0.94

 thereof

0.89

 relating

0.85

 imaginable

0.81

cale

0.80

 whilst

0.78

ources

0.76

pace

0.75

 arising

0.75

Activations Density 0.387%

words related to specific tasks or objects within a larger context

specific nouns related to various contexts, including music, data, questions, and evidence

No Comments

No Known Activations