INDEX

Explanations

terms related to specific social groups or environments

oai_token-act-pair · gpt-3.5-turbo

references to social or community groups

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

ãĥ¤

-0.70

 defe

-0.69

ensional

-0.68

 Cosponsors

-0.65

ression

-0.65

 codec

-0.64

iary

-0.64

eele

-0.63

ressor

-0.63

natureconservancy

-0.59

POSITIVE LOGITS

 circles

1.13

creen

0.96

 circle

0.87

pace

0.86

circle

0.85

jer

0.84

hift

0.83

naire

0.83

 circling

0.81

cale

0.81

Activations Density 0.010%

terms related to specific social groups or environments

references to social or community groups

No Comments

No Known Activations