INDEX

Explanations

statements where denial or refutation of allegations is expressed

oai_token-act-pair · gpt-3.5-turbo

statements of denial regarding involvement or wrongdoing

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Duration

-0.81

tnc

-0.76

 tremend

-0.72

rior

-0.69

icult

-0.67

omew

-0.67

inki

-0.67

ļéĨĴ

-0.66

ifully

-0.64

itiveness

-0.64

POSITIVE LOGITS

 wrongdoing

1.36

 involvement

1.05

 anything

1.04

nor

1.00

 malice

0.99

any

0.90

 allegation

0.88

 collusion

0.87

 improper

0.86

 interfered

0.84

Activations Density 0.230%

statements where denial or refutation of allegations is expressed

statements of denial regarding involvement or wrongdoing

No Comments

No Known Activations