INDEX

Explanations

punctuations that denote quotes or speech

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

pragma

-0.43

 pena

-0.37

 Wire

-0.35

ßig

-0.33

heritance

-0.33

rian

-0.33

存于互联网档案馆

-0.33

 بيها

-0.32

 fail

-0.32

Wire

-0.32

POSITIVE LOGITS

 betweenstory

0.63

 lenker

0.61

thâu

0.60

 resourceCulture

0.58

AutoScaleMode

0.58

رشف

0.57

 doInBackground

0.56

хьтан

0.56

WriteTagHelper

0.56

Personendaten

0.55

Activations Density 0.060%

punctuations that denote quotes or speech

No Comments

No Known Activations