INDEX

Explanations

references to the Apache organization

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.20.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

↵↵

-0.50

-0.49

-0.47

<eos>

-0.46

//

-0.44

er

-0.43

↵

-0.42

<bos>

-0.42

hline

-0.40

...

-0.40

POSITIVE LOGITS

 desmotivaciones

0.83

⪜

0.82

ésultats

0.82

redients

0.79

 témoig

0.77

majánló

0.76

 imagui

0.74

ſcher

0.73

 queſta

0.71

 FontWeight

0.71

Activations Density 0.002%

references to the Apache organization

No Comments

No Known Activations