INDEX

Explanations

phrases indicating a comparison or judgment based on some form of evidence

oai_token-act-pair · gpt-3.5-turbo

phrases that reference evidence or sources of information

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

arsity

-0.76

erella

-0.69

eer

-0.69

auer

-0.67

eon

-0.66

peat

-0.65

lac

-0.65

ERN

-0.64

ront

-0.64

rio

-0.64

POSITIVE LOGITS

 testimonies

0.88

ventory

0.85

 anecd

0.83

reports

0.82

 reports

0.80

 indications

0.76

 anecdotal

0.74

 hindsight

0.73

 findings

0.70

 observations

0.69

Activations Density 0.136%

phrases indicating a comparison or judgment based on some form of evidence

phrases that reference evidence or sources of information

No Comments

No Known Activations