INDEX

Explanations

words related to negative outcomes or events

oai_token-act-pair · gpt-3.5-turbo

terms related to disastrous events or situations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

trak

-0.85

etch

-0.80

bors

-0.77

ramid

-0.77

uni

-0.75

pel

-0.75

gat

-0.75

hung

-0.74

agos

-0.72

annis

-0.72

POSITIVE LOGITS

 havoc

1.00

itous

0.98

ãĥ¼ãĥĨ

0.89

 disastrous

0.85

 adolesc

0.79

ly

0.78

ously

0.75

ãĥ¥

0.74

 catastrophic

0.71

 consequences

0.71

Activations Density 0.009%

words related to negative outcomes or events

terms related to disastrous events or situations

No Comments

No Known Activations