INDEX

Explanations

phrases indicating location or direction

oai_token-act-pair · gpt-3.5-turbo

references to personal pronouns and expressions of intention or desire

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

naires

-0.86

advertisement

-0.65

classic

-0.64

oret

-0.63

ANN

-0.62

 Leilan

-0.60

ucci

-0.59

 ..............

-0.59

conn

-0.59

dra

-0.59

POSITIVE LOGITS

 nearest

1.07

 geographically

0.90

 weakest

0.87

 belong

0.81

 reside

0.79

 resided

0.79

 intersect

0.79

 originated

0.78

 belonged

0.77

 resides

0.76

Activations Density 0.172%

phrases indicating location or direction

references to personal pronouns and expressions of intention or desire

No Comments

No Known Activations