INDEX

Explanations

reflexive pronouns indicating self-referential actions or states

oai_token-act-pair · gpt-3.5-turbo

references to personal responsibility and self-agency

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

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

emis

-0.74

olid

-0.69

rought

-0.69

rise

-0.66

ammy

-0.65

iard

-0.63

wire

-0.62

 Federation

-0.62

heny

-0.62

yssey

-0.61

POSITIVE LOGITS

é¾įåĸļå£«

0.83

 mate

0.79

selves

0.75

ortium

0.71

ternally

0.67

æµ

0.66

sen

0.65

destruct

0.65

terday

0.65

 creatively

0.64

Activations Density 0.055%

reflexive pronouns indicating self-referential actions or states

references to personal responsibility and self-agency

No Comments

No Known Activations