INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

 Chains

-0.85

itive

-0.72

rr

-0.71

itives

-0.69

cific

-0.66

itutes

-0.65

relevant

-0.64

fits

-0.63

itionally

-0.63

IENT

-0.63

POSITIVE LOGITS

 roadway

0.70

wagen

0.68

 illum

0.67

luaj

0.65

 Columbus

0.63

 afore

0.63

Pen

0.62

editor

0.61

 writ

0.60

 highway

0.60

Activations Density 0.000%

No Known Activations

This feature has no known activations.