INDEX

Explanations

countries or geographical locations

oai_token-act-pair · gpt-3.5-turbo

references to locations and names associated with academic institutions or technical terms

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

ĸļ

-0.69

Jaw

-0.63

 McMaster

-0.55

 Gallagher

-0.54

 climbers

-0.54

enthal

-0.53

 retrospective

-0.51

EEE

-0.50

 occup

-0.50

isphere

-0.50

POSITIVE LOGITS

Cola

0.75

taboola

0.65

 Invalid

0.60

dayName

0.59

 Alto

0.57

aign

0.55

 ..............

0.55

=-=-

0.54

hazard

0.53

Chem

0.53

Activations Density 1.290%

countries or geographical locations

references to locations and names associated with academic institutions or technical terms

No Comments

No Known Activations