INDEX

Explanations

elements related to sentencing and language structure

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 12-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_12/width_16k/average_l0_82

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.12.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

 Warburton

-0.81

первых

-0.73

dolu

-0.73

ofire

-0.71

 חיצוניים

-0.70

icoot

-0.69

chaikovsky

-0.67

Warna

-0.66

 nicio

-0.66

AutoScaleMode

-0.65

POSITIVE LOGITS

 sentences

1.59

 sentence

1.54

 Sentence

1.51

 Sentences

1.36

sentences

1.31

Sentence

1.27

sentence

1.22

 sentenced

1.03

 frase

0.99

 Cune

0.94

Activations Density 0.136%

elements related to sentencing and language structure

No Comments

No Known Activations