INDEX

Explanations

pronouns related to personal identity

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "themselves."

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

amia

-0.73

ulton

-0.72

 Sierra

-0.72

ammy

-0.71

grade

-0.69

pour

-0.67

yson

-0.67

 Fulton

-0.67

emis

-0.66

MID

-0.64

POSITIVE LOGITS

selves

1.10

 tremend

0.95

 selves

0.88

self

0.85

 underwater

0.78

 conduc

0.78

 ashamed

0.76

 submar

0.75

 proport

0.75

 creatively

0.74

Activations Density 0.039%

pronouns related to personal identity

occurrences of the word "themselves."

No Comments

No Known Activations