INDEX

Explanations

discussions around processes and mechanisms, particularly in a cognitive or procedural context

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

 оригіналу

-0.69

IndentedString

-0.63

hlander

-0.53

ISupport

-0.51

 calendriers

-0.51

stasia

-0.51

 nakalista

-0.51

ēm

-0.51

DispatchToProps

-0.48

]]

-0.47

POSITIVE LOGITS

 through

1.11

 whereby

1.08

through

0.99

 Through

0.98

Through

0.98

by

0.94

 mediante

0.92

 THROUGH

0.89

 através

0.88

 durch

0.85

Activations Density 0.207%

discussions around processes and mechanisms, particularly in a cognitive or procedural context

No Comments

No Known Activations