INDEX

Explanations

phrases starting with "to" indicating intention or actions

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

)++;

-0.80

 Cæsar

-0.78

 Acapulco

-0.73

ſelves

-0.72

<bos>

-0.72

 Conscience

-0.72

 doubtnut

-0.72

 pleaſure

-0.71

]--;

-0.71

SOUNDBITE

-0.71

POSITIVE LOGITS

 gonna

1.32

 Gonna

1.12

 GONNA

1.04

gonna

1.00

Gonna

0.99

 going

0.82

 will

0.81

going

0.70

 Will

0.69

Will

0.66

Activations Density 0.100%

phrases starting with "to" indicating intention or actions

No Comments

No Known Activations