INDEX

Explanations

references to traffic-related concepts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 0-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_0/width_16k/average_l0_105

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.0.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

LabelTagHelper

-0.69

Xp

-0.66

 Camilo

-0.66

 piú

-0.65

 Demet

-0.63

dew

-0.63

 Pantheon

-0.62

whereas

-0.62

 Stolz

-0.62

 fett

-0.61

POSITIVE LOGITS

 traffic

0.77

ality

0.76

Non

0.71

non

0.71

traffic

0.67

Non

0.64

 Traffic

0.63

Maj

0.60

Traffic

0.59

in

0.58

Activations Density 0.187%

references to traffic-related concepts

No Comments

No Known Activations