INDEX

Explanations

words related to political discourse and criticism

oai_token-act-pair · gpt-3.5-turbo

critiques of political behavior and accountability

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

iple

-0.89

fm

-0.74

mx

-0.72

ibles

-0.70

vr

-0.70

DX

-0.68

VERTISEMENT

-0.68

isode

-0.68

uce

-0.66

rieving

-0.65

POSITIVE LOGITS

 namely

1.25

 irrespective

1.20

 thereby

1.20

 undermining

1.14

 lest

1.13

 ignoring

1.13

 whereas

1.12

 notwithstanding

1.11

 regardless

1.09

 perpetrated

1.08

Activations Density 0.394%

words related to political discourse and criticism

critiques of political behavior and accountability

No Comments

No Known Activations