INDEX

Explanations

words related to political processes

oai_token-act-pair · gpt-3.5-turbo

phrases related to public decision-making or referendums

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

bear

-0.63

 dear

-0.60

 Shack

-0.60

Yam

-0.60

 Seasons

-0.60

 Ares

-0.59

 Hels

-0.57

STER

-0.56

 Ingram

-0.56

Dispatch

-0.56

POSITIVE LOGITS

uits

1.22

uity

1.10

onduct

1.09

uit

1.08

rete

1.05

hetti

1.05

reen

1.03

aution

1.00

ussion

1.00

ount

1.00

Activations Density 0.038%

words related to political processes

phrases related to public decision-making or referendums

No Comments

No Known Activations