INDEX

Explanations

mentions of discussions or exchanges of words between individuals or groups

oai_token-act-pair · gpt-3.5-turbo

terms related to conversation or discussion

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

rug

-0.79

cot

-0.75

innon

-0.73

apolis

-0.73

purpose

-0.72

abol

-0.72

quer

-0.71

ramid

-0.71

roe

-0.70

rush

-0.69

POSITIVE LOGITS

 dialogue

1.50

 dialog

1.14

 Dialogue

1.13

 banter

0.95

ogue

0.95

 conversations

0.92

 conversation

0.91

ngth

0.89

 negotiation

0.89

 voice

0.82

Activations Density 0.006%

mentions of discussions or exchanges of words between individuals or groups

terms related to conversation or discussion

No Comments

No Known Activations