INDEX

Explanations

statements where individuals claim something

oai_token-act-pair · gpt-3.5-turbo

instances of the word "claimed."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

omb

-0.71

arrow

-0.65

 arrang

-0.65

 Closed

-0.63

gone

-0.62

schild

-0.62

ombo

-0.62

oyal

-0.62

 guiActiveUnfocused

-0.61

 Watching

-0.60

POSITIVE LOGITS

 innocence

0.85

 falsely

0.81

 ignorance

0.75

edly

0.74

 asylum

0.73

 responsibility

0.73

WTC

0.72

 superiority

0.69

 respons

0.68

 ownership

0.68

Activations Density 0.056%

statements where individuals claim something

instances of the word "claimed."

No Comments

No Known Activations