INDEX

Explanations

instances of objects or spaces being physically filled or packed with material

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "stuffed" and related concepts

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

SOURCE

-0.69

appro

-0.65

EFF

-0.64

 diligence

-0.63

mentioned

-0.63

relations

-0.63

externalActionCode

-0.62

MAL

-0.62

APH

-0.62

INAL

-0.61

POSITIVE LOGITS

 stuffed

1.14

nesday

0.96

bags

0.90

bag

0.88

 stuffing

0.85

ulously

0.83

 toys

0.83

 plush

0.83

ertodd

0.82

 sacks

0.80

Activations Density 0.009%

instances of objects or spaces being physically filled or packed with material

occurrences of the word "stuffed" and related concepts

No Comments

No Known Activations