INDEX

Explanations

references to signatures and related terms

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 4-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_4/width_16k/average_l0_124

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.4.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

apollo

-0.55

oob

-0.54

subsubsection

-0.54

vx

-0.54

ensk

-0.54

enedor

-0.52

 Abbott

-0.51

>*/

-0.51

Tel

-0.51

POSITIVE LOGITS

 signature

3.00

 Signature

2.75

 signatures

2.74

 Signatures

2.64

signature

2.63

Signature

2.45

Signatures

2.30

 SIGNATURE

2.30

signatures

2.29

SIGNATURE

2.04

Activations Density 0.092%

references to signatures and related terms

No Comments

No Known Activations