INDEX

Explanations

proper nouns related to politics

oai_token-act-pair · gpt-3.5-turbo

mentions of specific individuals, particularly those named Gloria and Rove

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

aws

-0.80

erella

-0.76

adoes

-0.76

icum

-0.74

ook

-0.74

UAL

-0.73

osuke

-0.73

ivated

-0.73

ipper

-0.73

ooters

-0.72

POSITIVE LOGITS

 Gloria

0.90

 Tone

0.80

verts

0.78

fol

0.78

holes

0.77

tex

0.74

beat

0.74

lla

0.71

jet

0.71

hole

0.71

Activations Density 0.015%

proper nouns related to politics

mentions of specific individuals, particularly those named Gloria and Rove

No Comments

No Known Activations