INDEX

Explanations

references to bars or bar-like structures

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "bar" 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

lihood

-0.81

UAL

-0.76

IBLE

-0.75

 Disorder

-0.75

 Languages

-0.71

 Scarlet

-0.71

 Outbreak

-0.71

CVE

-0.70

EEE

-0.69

ãģį

-0.68

POSITIVE LOGITS

itone

1.32

celona

1.27

bell

1.16

bara

1.10

iatric

1.05

becue

0.96

bed

0.95

riers

0.93

bour

0.92

rison

0.90

Activations Density 0.018%

references to bars or bar-like structures

occurrences of the word "bar" in various contexts

No Comments

No Known Activations