INDEX

Explanations

references to file names and related attributes in documents

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 11-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_11/width_16k/average_l0_80

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.11.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

-0.61

Protein

-0.56

 parlé

-0.52

 pubblici

-0.52

 Protein

-0.52

 protein

-0.51

protein

-0.51

 Ambrose

-0.48

 природе

-0.45

POSITIVE LOGITS

 filename

1.74

filename

1.55

Filename

1.14

 Filename

1.07

 filenames

1.04

matchCondition

1.00

FILENAME

0.99

 Chwiliwch

0.92

filenames

0.88

 ModelRenderer

0.88

Activations Density 0.037%

references to file names and related attributes in documents

No Comments

No Known Activations