INDEX

Explanations

words and phrases related to roles and classifications within various groups

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 25-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_25/width_16k/average_l0_116

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.25.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

 cherchés

-0.71

were

-0.64

Were

-0.58

 gdyby

-0.57

Has

-0.55

extAlignment

-0.55

Has

-0.54

 Were

-0.53

 theyre

-0.52

سمبر

-0.51

POSITIVE LOGITS

 love

0.92

 often

0.88

 rarely

0.81

 tend

0.81

 seldom

0.76

 spend

0.75

 typically

0.75

 learn

0.71

 shouldn

0.69

 LOVE

0.69

Activations Density 0.497%

words and phrases related to roles and classifications within various groups

No Comments

No Known Activations