INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

 Steel

-0.64

gig

-0.64

apest

-0.64

 Gleaming

-0.64

ESP

-0.61

 Polk

-0.61

AU

-0.61

================================================================

-0.60

imon

-0.60

 Bundesliga

-0.60

POSITIVE LOGITS

udence

0.78

iding

0.76

rahim

0.75

engers

0.73

ults

0.72

rogram

0.72

ptives

0.70

habi

0.70

hend

0.69

HAEL

0.69

Activations Density 0.000%

No Known Activations

This feature has no known activations.