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

fish

-0.73

apo

-0.69

ails

-0.68

ounded

-0.67

oths

-0.65

hra

-0.64

horn

-0.63

hap

-0.62

erves

-0.61

 vanquished

-0.61

POSITIVE LOGITS

Work

0.71

 informational

0.71

âĨĳ

0.69

mson

0.65

WARNING

0.64

pmwiki

0.63

[*

0.62

 WARN

0.61

AMA

0.61

 redundancy

0.60

Activations Density 0.000%

No Known Activations

This feature has no known activations.