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

culosis

-0.83

ographed

-0.70

 stim

-0.63

 Passing

-0.62

 passing

-0.61

 Strikes

-0.61

 inoc

-0.61

 scrim

-0.61

 borne

-0.60

IFIED

-0.60

POSITIVE LOGITS

Query

0.77

ql

0.76

SQL

0.74

selage

0.70

zers

0.70

 isEnabled

0.68

 Methodist

0.67

adata

0.66

Meta

0.66

onen

0.66

Activations Density 0.000%

No Known Activations

This feature has no known activations.