INDEX

Explanations

mentions of requests or instances of extending something

oai_token-act-pair · gpt-3.5-turbo

mentions of the term "extension" 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

ivan

-0.83

iard

-0.79

icial

-0.78

\":

-0.78

ards

-0.77

ãĥ£

-0.75

bold

-0.72

lins

-0.72

illard

-0.68

cca

-0.68

POSITIVE LOGITS

 cords

1.02

 extensions

0.92

 extension

0.89

 cord

0.89

 ende

0.84

 servicing

0.77

 therap

0.77

 applic

0.75

 extending

0.73

 adolesc

0.73

Activations Density 0.017%

mentions of requests or instances of extending something

mentions of the term "extension" in various contexts

No Comments

No Known Activations