INDEX

Explanations

mentions of wholeness or completeness

oai_token-act-pair · gpt-3.5-turbo

references to the word "entire" in various contexts

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

friends

-0.75

none

-0.71

chairs

-0.71

killers

-0.70

shows

-0.68

acons

-0.66

rants

-0.65

ju

-0.65

KER

-0.64

 whispers

-0.63

POSITIVE LOGITS

 length

0.88

 spectrum

0.86

 continent

0.86

 ordeal

0.86

hearted

0.84

 globe

0.83

gam

0.82

 arsenal

0.80

 duration

0.79

heartedly

0.76

Activations Density 0.037%

mentions of wholeness or completeness

references to the word "entire" in various contexts

No Comments

No Known Activations