INDEX

Explanations

No Explanations Found

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

 Chevy

-0.72

ller

-0.72

MLG

-0.71

erella

-0.69

culosis

-0.67

vati

-0.66

 Franch

-0.66

 Optimus

-0.64

Pwr

-0.64

 Hann

-0.64

POSITIVE LOGITS

achev

0.77

ĨĴ

0.73

?????-

0.71

 prag

0.69

 urgency

0.69

 semantics

0.68

ĪĴ

0.68

Dispatch

0.67

Response

0.67

Tokens

0.64

Activations Density 0.000%

No Known Activations

This feature has no known activations.