INDEX

Explanations

references to guides and interviews that demonstrate positive evaluations or experiences

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

 MainAxisSize

-0.42

 AssemblyProduct

-0.40

fastjson

-0.35

 getItemId

-0.35

awtextra

-0.35

MENAFN

-0.34

engelsk

-0.34

 sistema

-0.34

ViewFeatures

-0.33

Sprintf

-0.33

POSITIVE LOGITS

itespace

0.52

 IBOutlet

0.52

IndentedString

0.50

 '\\;'

0.47

 tocó

0.47

chose

0.47

intele

0.47

 chose

0.46

ske

0.45

 Carriers

0.45

Activations Density 0.005%

references to guides and interviews that demonstrate positive evaluations or experiences

No Comments

No Known Activations