INDEX

Explanations

names of people or places

oai_token-act-pair · gpt-3.5-turbo

proper names and significant terms in a narrative context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

Reviewer

-0.58

ãĤ¢ãĥ«

-0.49

ãĤ®

-0.48

éĹĺ

-0.47

inarily

-0.45

rongh

-0.44

ividual

-0.44

âĢ¢âĢ¢

-0.43

 coincide

-0.43

ãĥķ

-0.43

POSITIVE LOGITS

endon

0.53

ias

0.51

minist

0.50

athing

0.49

terness

0.49

oli

0.48

puff

0.47

antis

0.46

esson

0.46

clerosis

0.46

Activations Density 2.742%

names of people or places

proper names and significant terms in a narrative context

No Comments

No Known Activations