INDEX

Explanations

numbers

oai_token-act-pair · gpt-3.5-turbo

specific numeric and alphanumeric identifiers related to items, locations, or sequences

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

hern

-0.70

 spirited

-0.66

 Entered

-0.64

 awoken

-0.60

sea

-0.60

 Hague

-0.59

 Grateful

-0.58

 wartime

-0.58

Merit

-0.57

ter

-0.55

POSITIVE LOGITS

ogether

0.71

rants

0.67

aza

0.64

ools

0.61

Tokens

0.61

inos

0.60

 veter

0.57

oice

0.55

 pods

0.55

radius

0.55

Activations Density 1.267%

numbers

specific numeric and alphanumeric identifiers related to items, locations, or sequences

No Comments

No Known Activations