INDEX

Explanations

mentions of a particular place or time within a given context

oai_token-act-pair · gpt-3.5-turbo

references to specific instances or events in discussions

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

isms

-0.72

cks

-0.70

ickets

-0.70

letes

-0.70

sers

-0.68

masters

-0.67

omers

-0.66

uties

-0.66

onto

-0.65

acers

-0.65

POSITIVE LOGITS

 regard

1.20

 vicinity

1.10

 vein

1.04

 context

1.04

 particular

0.98

 week

0.95

 manner

0.95

 guise

0.91

 predicament

0.90

 area

0.89

Activations Density 0.061%

mentions of a particular place or time within a given context

references to specific instances or events in discussions

No Comments

No Known Activations