INDEX

Explanations

references to coordinate systems or positions in graphical contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 5-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_5/width_16k/average_l0_68

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.5.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

 himself

-0.71

ThroughAttribute

-0.65

 lenker

-0.61

 sand

-0.61

ⓧ

-0.58

 prostate

-0.56

 manhood

-0.55

himself

-0.54

 estimés

-0.54

POSITIVE LOGITS

yAxis

1.03

 ymax

0.86

ylabel

0.85

yaxis

0.82

 yPos

0.82

yticks

0.78

minY

0.78

clientY

0.76

 giggled

0.76

ypos

0.75

Activations Density 0.614%

references to coordinate systems or positions in graphical contexts

No Comments

No Known Activations