INDEX

Explanations

temporal adverbs indicating past events or changes in circumstances

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-2b-pt-res/layer_2/width_16k/average_l0_141

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.2.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

endphp

-0.85

 Monfieur

-0.83

cupertino

-0.77

 poffible

-0.71

UnusedPrivate

-0.71

 forceps

-0.71

 calms

-0.70

 Completion

-0.68

 polarity

-0.68

 pleaſure

-0.67

POSITIVE LOGITS

 often

0.89

 always

0.83

 routinely

0.69

 still

0.67

 also

0.67

 repeatedly

0.65

 frequently

0.64

often

0.63

 previously

0.63

 effectively

0.62

Activations Density 0.731%

temporal adverbs indicating past events or changes in circumstances

No Comments

No Known Activations