INDEX

Explanations

specific contexts or situations

oai_token-act-pair · gpt-3.5-turbo

phrases that indicate context or comparison

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

rib

-0.61

arna

-0.59

 Regist

-0.58

NetMessage

-0.58

ERE

-0.56

terday

-0.55

orsi

-0.55

 conclud

-0.55

ank

-0.55

rosso

-0.54

POSITIVE LOGITS

forts

0.84

owitz

0.69

achu

0.68

cription

0.63

liest

0.62

verse

0.61

kees

0.61

pires

0.60

lights

0.60

osphere

0.59

Activations Density 0.217%

specific contexts or situations

phrases that indicate context or comparison

No Comments

No Known Activations