INDEX

Explanations

phrases that suggest purpose or intent

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

 Theſe

-0.90

 ujednoznacz

-0.81

sizeCache

-0.81

 MainAxisSize

-0.80

 ſche

-0.79

 resourceCulture

-0.79

 ſeveral

-0.78

 ſever

-0.78

 vixion

-0.77

 fubject

-0.77

POSITIVE LOGITS

 pour

1.04

for

0.96

por

0.92

 για

0.89

For

0.88

 para

0.84

Για

0.83

 för

0.83

 pentru

0.81

 Untuk

0.77

Activations Density 0.053%

phrases that suggest purpose or intent

No Comments

No Known Activations