INDEX

Explanations

references to investigations or examinations being conducted

oai_token-act-pair · gpt-3.5-turbo

terms related to official investigations or inquiries

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

 Extrem

-0.75

AMES

-0.68

Decl

-0.68

Aid

-0.66

IAN

-0.65

 compe

-0.64

vari

-0.62

 reconc

-0.61

relative

-0.61

activated

-0.61

POSITIVE LOGITS

 probe

1.14

 Probe

1.10

 probes

1.09

 probing

1.08

 inquiry

0.86

agher

0.84

 investigating

0.76

 investigation

0.76

 gauge

0.74

 questioning

0.72

Activations Density 0.010%

references to investigations or examinations being conducted

terms related to official investigations or inquiries

No Comments

No Known Activations