INDEX

Explanations

beginning or initiation phrases in sentences

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 7-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_7/width_16k/average_l0_69

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.7.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

 research

-0.54

WithMany

-0.53

 people

-0.52

רים

-0.51

/***/

-0.49

sub

-0.49

 text

-0.48

 types

-0.48

 type

-0.48

iers

-0.47

POSITIVE LOGITS

 propOrder

0.94

setVerticalGroup

0.77

 <=",

0.77

новниш

0.76

 himſelf

0.75

IntoConstraints

0.74

 myſelf

0.74

 themſelves

0.73

RegressionTest

0.73

expandindo

0.73

Activations Density 0.245%

beginning or initiation phrases in sentences

No Comments

No Known Activations