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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

pta

-0.92

acters

-0.74

 Bucc

-0.73

ptives

-0.68

qua

-0.68

ounters

-0.68

 Wink

-0.68

uras

-0.67

ieri

-0.67

eele

-0.66

POSITIVE LOGITS

 aggregation

0.69

 subcontract

0.67

 athletic

0.65

KA

0.64

 aggreg

0.63

nai

0.62

 aggregate

0.61

 resettlement

0.60

 arsen

0.60

 Eater

0.59

Activations Density 0.000%

No Known Activations

This feature has no known activations.