INDEX

Explanations

mentions of thin objects or materials

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "thin" in various contexts

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

ILLE

-0.74

OUP

-0.70

udeau

-0.70

pheus

-0.69

NL

-0.68

Particip

-0.68

andise

-0.67

 Pact

-0.67

tumblr

-0.67

 Volt

-0.66

POSITIVE LOGITS

ned

1.31

ning

1.16

ners

1.13

nesses

1.00

 thin

0.88

ened

0.86

 slices

0.85

ther

0.85

ening

0.85

odynamic

0.85

Activations Density 0.013%

mentions of thin objects or materials

occurrences of the word "thin" in various contexts

No Comments

No Known Activations