INDEX

Explanations

terms related to file extensions

oai_token-act-pair · gpt-3.5-turbo

mentions of "extension" related to various contexts

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

\":

-0.81

iard

-0.78

ãĥ£

-0.75

ivan

-0.74

ky

-0.73

kie

-0.73

icial

-0.73

yah

-0.70

avery

-0.68

POSITIVE LOGITS

 cords

1.06

 cord

0.95

 fuse

0.76

 thereof

0.73

 extensions

0.73

 extension

0.69

 wrench

0.68

liga

0.67

ibility

0.67

 servicing

0.66

Activations Density 0.029%

terms related to file extensions

mentions of "extension" related to various contexts

No Comments

No Known Activations