INDEX

Explanations

references to suicide-related terms

oai_token-act-pair · gpt-3.5-turbo

references to suicide and related topics

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

rium

-0.76

asonic

-0.73

heny

-0.72

oyal

-0.71

artisan

-0.71

aunder

-0.70

Avg

-0.70

parts

-0.66

bred

-0.66

 Correct

-0.65

POSITIVE LOGITS

 suicide

1.16

 bomber

1.04

icide

1.04

zai

1.02

icides

0.99

 bombers

0.94

itated

0.91

itating

0.90

 suicides

0.84

ide

0.82

Activations Density 0.011%

references to suicide-related terms

references to suicide and related topics

No Comments

No Known Activations