INDEX

Explanations

phrases or words indicating a similarity or correspondence

oai_token-act-pair · gpt-3.5-turbo

phrases related to matching or compatibility

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

gins

-0.73

 trave

-0.71

seeking

-0.68

seek

-0.68

ft

-0.67

cember

-0.67

deal

-0.67

ahs

-0.67

worms

-0.66

uler

-0.66

POSITIVE LOGITS

 ours

0.89

 hers

0.78

 precon

0.76

peak

0.73

 Tyrann

0.71

 perfectly

0.70

 neatly

0.68

 pree

0.67

 closely

0.66

 coincide

0.65

Activations Density 0.113%

phrases or words indicating a similarity or correspondence

phrases related to matching or compatibility

No Comments

No Known Activations