INDEX

Explanations

occurrences of the word "iter" and its variations, indicating a focus on iterations and related programming concepts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 1-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_1/width_16k/average_l0_69

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.1.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

expandindo

-0.42

 Ferris

-0.39

 alcalde

-0.38

SEGUIR

-0.34

 enseña

-0.33

将军

-0.33

hemiah

-0.32

 cuales

-0.32

 poveznice

-0.32

 Venedig

-0.32

POSITIVE LOGITS

TER

1.06

ter

1.02

tering

0.99

iter

0.95

ters

0.91

tered

0.91

ter

0.84

tery

0.83

eter

0.82

tera

0.78

Activations Density 0.326%

occurrences of the word "iter" and its variations, indicating a focus on iterations and related programming concepts

No Comments

No Known Activations