INDEX

Explanations

words related to political events and conflicts

oai_token-act-pair · gpt-3.5-turbo

references to political events or government actions

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Ń·

-0.80

erity

-0.74

Down

-0.72

Īè

-0.71

=-=-=-=-=-=-=-=-

-0.66

ials

-0.64

orthern

-0.64

ãĤ®

-0.63

 Gladiator

-0.63

çĶ°

-0.62

POSITIVE LOGITS

Cue

0.69

AE

0.68

stellar

0.65

cum

0.64

gag

0.62

dom

0.62

hack

0.62

Boo

0.61

cade

0.61

Cf

0.60

Activations Density 0.000%

words related to political events and conflicts

references to political events or government actions

No Comments

No Known Activations