INDEX

Explanations

proper nouns related to media and entertainment

oai_token-act-pair · gpt-3.5-turbo

titles of specific media works such as movies, shows, or news articles

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

ruary

-0.84

terday

-0.75

avorite

-0.70

sed

-0.66

opausal

-0.64

 capacities

-0.63

antly

-0.62

aukee

-0.62

 mitigation

-0.61

ierrez

-0.60

POSITIVE LOGITS

 Awakens

0.97

 Angels

0.82

isky

0.79

 Battalion

0.75

 Trilogy

0.74

arth

0.74

 Files

0.74

 Bride

0.73

 Herald

0.72

 soundtrack

0.71

Activations Density 0.294%

proper nouns related to media and entertainment

titles of specific media works such as movies, shows, or news articles

No Comments

No Known Activations