INDEX

Explanations

phrases that convey varying degrees of interpretation or contextual understanding

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

 astore

-0.69

legte

-0.61

AsUp

-0.56

λιο

-0.53

 Webb

-0.53

 automatica

-0.52

 thẳng

-0.51

分别是

-0.50

lez

-0.50

gle

-0.50

POSITIVE LOGITS

 تضيفلها

0.76

 Extent

0.70

LabelTagHelper

0.69

 extent

0.69

yyr

0.67

 становника

0.66

揄

0.65

)":

0.64

مق

0.64

 Hinsicht

0.64

Activations Density 0.279%

phrases that convey varying degrees of interpretation or contextual understanding

No Comments

No Known Activations