INDEX

Explanations

words related to challenging or opposing others

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "confront."

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

rotein

-0.76

exempt

-0.75

ectar

-0.74

adder

-0.74

protein

-0.72

flu

-0.72

serving

-0.71

sample

-0.71

RNA

-0.71

Ĥ¬

-0.69

POSITIVE LOGITS

 confront

1.20

 confrontation

0.93

 confronting

0.92

 confronts

0.87

ational

0.87

DonaldTrump

0.82

 Tanz

0.79

ings

0.78

 encounters

0.77

 truths

0.77

Activations Density 0.006%

words related to challenging or opposing others

occurrences of the word "confront."

No Comments

No Known Activations