INDEX

Explanations

references to images with accompanying text descriptions

oai_token-act-pair · gpt-3.5-turbo

the presence of the word "Hide" in the context of content presentation

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

etheless

-0.85

eele

-0.78

ammy

-0.77

ilingual

-0.77

odic

-0.74

ounty

-0.73

rontal

-0.70

issance

-0.70

enegger

-0.68

 confir

-0.68

POSITIVE LOGITS

 Caption

1.08

away

0.98

Hide

0.93

 Hide

0.89

ously

0.85

Streamer

0.74

hide

0.74

Pic

0.71

Emb

0.70

Track

0.70

Activations Density 0.014%

references to images with accompanying text descriptions

the presence of the word "Hide" in the context of content presentation

No Comments

No Known Activations