INDEX

Explanations

elements related to structured documentation or code

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 21-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_21/width_16k/average_l0_129

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.21.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

 queſta

-1.09

 laſſen

-0.97

 ſind

-0.96

ロウィン

-0.95

ſſung

-0.95

 müſſen

-0.94

iſchen

-0.94

iſche

-0.94

ſchaft

-0.93

ſicht

-0.93

POSITIVE LOGITS

hline

0.65

↵↵

0.57

0.56

0.54

In

0.49

if

0.47

So

0.46

0.45

Activations Density 0.048%

elements related to structured documentation or code

No Comments

No Known Activations