INDEX

Explanations

mentions of names, likely linked to politics or news events

oai_token-act-pair · gpt-3.5-turbo

instances of specific names relevant to political figures

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

chnology

-0.84

ppel

-0.77

vern

-0.72

ppy

-0.70

 pals

-0.69

chrom

-0.67

arest

-0.65

erest

-0.64

MENT

-0.64

utsche

-0.64

POSITIVE LOGITS

ibn

1.25

 Ahmad

1.22

Ali

1.20

 Hassan

1.20

 Sheikh

1.18

iyah

1.18

 Jamal

1.12

 Mohamed

1.10

 Muhammad

1.08

abad

1.08

Activations Density 0.051%

mentions of names, likely linked to politics or news events

instances of specific names relevant to political figures

No Comments

No Known Activations