INDEX

Explanations

mentions of specific movies or entertainment productions

oai_token-act-pair · gpt-3.5-turbo

references to animated films or properties

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

 kidneys

-0.73

merce

-0.72

aid

-0.71

 GOODMAN

-0.69

 booming

-0.69

 Forrest

-0.68

 contrace

-0.65

 disciplines

-0.65

ifted

-0.65

 comma

-0.64

POSITIVE LOGITS

pic

1.27

amera

1.03

Pic

1.00

colo

0.99

uous

0.97

uously

0.95

nic

0.92

nick

0.90

ulous

0.89

iple

0.88

Activations Density 0.009%

mentions of specific movies or entertainment productions

references to animated films or properties

No Comments

No Known Activations