INDEX

Explanations

phrases indicating addition or continuation in a sentence

oai_token-act-pair · gpt-3.5-turbo

instances of reported speech or citations

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

peg

-0.79

tes

-0.67

unker

-0.66

orce

-0.64

 Stronghold

-0.63

cus

-0.63

uddy

-0.59

pires

-0.59

course

-0.59

Ã»

-0.57

POSITIVE LOGITS

apest

0.69

uations

0.67

 sarc

0.63

 sarcast

0.63

[+

0.62

iT

0.62

 quoting

0.61

amera

0.60

BUS

0.60

ribune

0.59

Activations Density 0.081%

phrases indicating addition or continuation in a sentence

instances of reported speech or citations

No Comments

No Known Activations