INDEX

Explanations

names 'Ron' with varying degrees of relevance

oai_token-act-pair · gpt-3.5-turbo

mentions of individuals named Ron

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

td

-0.67

hower

-0.65

 Chiefs

-0.62

 Annotations

-0.62

 Pixie

-0.62

 sugg

-0.61

 completion

-0.59

 Dakota

-0.59

LOAD

-0.58

displayText

-0.57

POSITIVE LOGITS

ald

0.98

aldo

0.95

ny

0.95

 Weasley

0.94

 Swanson

0.93

nie

0.91

igans

0.84

Paul

0.84

bie

0.83

nen

0.83

Activations Density 0.013%

names 'Ron' with varying degrees of relevance

mentions of individuals named Ron

No Comments

No Known Activations