INDEX

Explanations

instances of phrases indicating a numerical quantity followed by a general noun or concept

oai_token-act-pair · gpt-3.5-turbo

phrases indicating a relative time or duration

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

 Hollow

-0.63

 Prelude

-0.61

 Cassandra

-0.59

 Emblem

-0.59

LF

-0.59

first

-0.58

-0.57

Osw

-0.56

alist

-0.56

 Romeo

-0.56

POSITIVE LOGITS

bered

0.90

oths

0.87

othe

0.86

zin

0.76

éĹ

0.75

oshenko

0.75

ricular

0.74

ooo

0.74

apy

0.73

posium

0.73

Activations Density 0.012%

instances of phrases indicating a numerical quantity followed by a general noun or concept

phrases indicating a relative time or duration

No Comments

No Known Activations