INDEX

Explanations

references to political events and entities

oai_token-act-pair · gpt-3.5-turbo

phrases related to political events and campaigns

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

 meanwhile

-0.44

 ducks

-0.44

roo

-0.43

awa

-0.43

baugh

-0.42

milo

-0.42

rily

-0.42

 faint

-0.41

 whales

-0.41

ensibly

-0.41

POSITIVE LOGITS

)).

0.49

 Elys

0.47

]).

0.47

Teg

0.47

arcer

0.45

]),

0.45

Aug

0.43

 Neuroscience

0.43

 msec

0.43

 Trib

0.42

Activations Density 2.127%

references to political events and entities

phrases related to political events and campaigns

No Comments

No Known Activations