INDEX

Explanations

the word "kind."

oai_token-act-pair · gpt-3.5-turbo

phrases that express a sense of mild inevitability or descriptor-like qualities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

oulos

-0.85

perty

-0.72

edia

-0.71

IBLE

-0.69

PT

-0.66

 VIDEOS

-0.66

ERY

-0.66

OPE

-0.65

å§«

-0.63

senal

-0.63

POSITIVE LOGITS

cast

0.89

ling

0.82

este

0.81

hearted

0.80

nered

0.79

erest

0.79

lier

0.74

assy

0.71

ends

0.69

erd

0.68

Activations Density 0.028%

the word "kind."

phrases that express a sense of mild inevitability or descriptor-like qualities

No Comments

No Known Activations