INDEX

Explanations

phrases indicating proximity or nearness

oai_token-act-pair · gpt-3.5-turbo

repeated phrases indicating proximity or closeness

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

birds

-0.65

 behav

-0.65

bet

-0.63

!!!!!!!!

-0.62

apo

-0.62

uploads

-0.60

NS

-0.59

nep

-0.59

RAW

-0.58

 corrid

-0.58

POSITIVE LOGITS

sighted

0.74

 heels

0.71

 scrutiny

0.68

entin

0.68

 sidx

0.66

othal

0.66

izabeth

0.65

amaru

0.65

ricular

0.65

 shave

0.65

Activations Density 0.117%

phrases indicating proximity or nearness

repeated phrases indicating proximity or closeness

No Comments

No Known Activations