INDEX

Explanations

programming functions and method definitions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 15-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_15/width_16k/average_l0_78

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.15.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

aarrggbb

-0.69

 transfieras

-0.65

+#+

-0.63

zések

-0.63

ר

-0.58

::*;

-0.58

 Trop

-0.57

>>;

-0.56

Sleepy

-0.55

 ModelExpression

-0.54

POSITIVE LOGITS

 שוליים

0.77

 FishBase

0.69

躇

0.69

</caption>

0.65

tvguidetime

0.62

COLN

0.59

 venons

0.59

\{\\

0.59

InjectAttribute

0.59

cierto

0.57

Activations Density 0.027%

programming functions and method definitions

No Comments

No Known Activations