INDEX

Explanations

repeated use of the term "considering."

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

 considered

-0.63

 part

-0.53

tid

-0.51

 Considered

-0.51

considered

-0.51

lig

-0.50

ache

-0.49

set

-0.49

ms

-0.48

POSITIVE LOGITS

considering

1.45

 considering

1.43

 Considering

1.35

Considering

1.32

 considerando

1.00

]--;

0.83

Assuming

0.77

+#+

0.75

]]);

0.74

 teniendo

0.74

Activations Density 0.119%

repeated use of the term "considering."

No Comments

No Known Activations