INDEX

Explanations

references to physical environments and settings

oai_token-act-pair · gpt-3.5-turbo

references to environments or settings

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

WAR

-0.64

kick

-0.62

cher

-0.59

quist

-0.59

 valid

-0.58

ict

-0.57

 either

-0.57

bit

-0.55

NOW

-0.55

one

-0.55

POSITIVE LOGITS

 surroundings

3.95

 vicinity

1.61

 surrounds

1.49

 surrounding

1.44

 scenery

1.37

 environments

1.36

 environment

1.36

vironments

1.23

 neighbours

1.20

 atmosphere

1.14

Activations Density 0.022%

references to physical environments and settings

references to environments or settings

No Comments

No Known Activations