INDEX

Explanations

political figures and affiliations

oai_token-act-pair · gpt-3.5-turbo

references to political figures and contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

string

-0.72

äºº

-0.66

å£«

-0.65

 sued

-0.63

 claims

-0.62

 certs

-0.62

 negligence

-0.61

 projectiles

-0.61

 rival

-0.61

 Unicode

-0.60

POSITIVE LOGITS

podcast

1.05

odcast

1.02

Interview

1.01

isode

0.99

aturday

0.93

Topics

0.88

 discussing

0.87

 interview

0.86

 podcast

0.86

 listener

0.86

Activations Density 0.848%

political figures and affiliations

references to political figures and contexts

No Comments

No Known Activations