INDEX

Explanations

phrases related to putting an end to something or casting doubt on a situation

oai_token-act-pair · gpt-3.5-turbo

phrases indicating conclusions, doubts, and efforts to resolve issues

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

 motor

-0.65

 quint

-0.63

 clichÃ©

-0.63

 Ascension

-0.59

 Horizons

-0.57

 notoriously

-0.57

 constant

-0.56

 Bugs

-0.56

 chants

-0.56

 eccentric

-0.56

POSITIVE LOGITS

 thereto

0.79

enance

0.75

 sidx

0.74

MpServer

0.72

=#

0.72

oise

0.71

Mich

0.71

çīĪ

0.71

pees

0.70

 Flavoring

0.70

Activations Density 0.167%

phrases related to putting an end to something or casting doubt on a situation

phrases indicating conclusions, doubts, and efforts to resolve issues

No Comments

No Known Activations