INDEX

Explanations

instances of the term "heads," particularly in contexts suggesting authority or leadership

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-2b-pt-res/layer_1/width_16k/average_l0_102

Prompts (Dashboard)

36,864 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

ślę

-0.65

msubsup

-0.63

prising

-0.57

Interop

-0.56

 paramInt

-0.56

 lemak

-0.53

glGen

-0.52

 obstante

-0.52

ibu

-0.51

ValueStyle

-0.50

POSITIVE LOGITS

 heads

1.27

Heads

1.22

 Heads

1.22

heads

1.02

 brains

0.91

 hearts

0.89

 minds

0.82

 Hearts

0.75

OGND

0.73

brains

0.72

Activations Density 0.108%

instances of the term "heads," particularly in contexts suggesting authority or leadership

No Comments

No Known Activations