INDEX

Explanations

specific patterns of letters or structures in code or text snippets

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

 myſelf

-0.85

 itſelf

-0.84

 Monfieur

-0.75

 Jefus

-0.71

ſelves

-0.69

 ſever

-0.68

CppMethod

-0.67

 whoſe

-0.66

ValueStyle

-0.66

 ſeveral

-0.66

POSITIVE LOGITS

 finger

0.53

finger

0.51

 Finger

0.50

 Bryan

0.49

()],

0.44

Finger

0.44

 незавершена

0.43

SystemColors

0.43

TableField

0.43

tagHelperRunner

0.43

Activations Density 0.116%

specific patterns of letters or structures in code or text snippets

No Comments

No Known Activations