INDEX

Explanations

phrases related to writing or executing scripts in a computer programming context

oai_token-act-pair · gpt-3.5-turbo

references to scripts and scripting-related terminology

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

Dew

-0.72

arthed

-0.70

 tide

-0.69

 Psychiat

-0.68

 Alto

-0.66

coat

-0.66

 Franch

-0.65

obby

-0.64

 Floor

-0.64

hoff

-0.63

POSITIVE LOGITS

ural

0.98

 scripting

0.95

scripts

0.93

 interpreter

0.93

Pub

0.90

Deploy

0.90

urally

0.88

script

0.88

 scripts

0.87

 script

0.87

Activations Density 0.010%

phrases related to writing or executing scripts in a computer programming context

references to scripts and scripting-related terminology

No Comments

No Known Activations