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

Sa

-0.77

weekly

-0.77

SPA

-0.77

PF

-0.75

oo

-0.74

aked

-0.74

Coming

-0.73

oe

-0.73

unique

-0.71

HC

-0.71

POSITIVE LOGITS

 Isles

0.73

 assorted

0.72

 thereby

0.72

 directs

0.71

 hence

0.69

 perhaps

0.67

 preferably

0.67

 thus

0.66

 intimid

0.66

 Ukrain

0.65

Activations Density 0.000%

No Known Activations

This feature has no known activations.