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

SPONSORED

-0.81

ãĥĺ

-0.78

ãĥį

-0.76

PDATE

-0.69

à¼

-0.68

Interested

-0.64

RAW

-0.63

IVES

-0.63

î

-0.61

STAT

-0.61

POSITIVE LOGITS

jong

0.77

jri

0.76

acter

0.73

haar

0.73

arin

0.73

berg

0.69

atz

0.69

stein

0.68

vu

0.66

Ã©e

0.66

Activations Density 0.000%

No Known Activations

This feature has no known activations.