INDEX

Explanations

sequences of words related to physical violence and harm

oai_token-act-pair · gpt-3.5-turbo

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

ĕ

-0.56

ù

-0.55

Ğ

-0.55

Ĉ

-0.55

č

-0.55

é¾įå¥

-0.55

Ă

-0.55

ý

-0.55

Č

-0.55

POSITIVE LOGITS

again

0.72

 ensued

0.69

 respectively

0.68

 altogether

0.67

*.

0.65

 versa

0.62

!.

0.61

 accordingly

0.60

 equivalents

0.60

 Again

0.60

Activations Density 0.977%

sequences of words related to physical violence and harm

No Comments

No Known Activations