INDEX

Explanations

mathematical constructs and notation

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 7-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_7/width_16k/average_l0_69

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.7.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

 BoxFit

-0.87

tahui

-0.75

بوابة

-0.68

HOY

-0.65

bited

-0.62

-0.61

Portály

-0.60

ciled

-0.60

+#+

-0.59

Linki

-0.59

POSITIVE LOGITS

[toxicity=0]

1.03

 ghiaccio

0.55

 okuyayım

0.52

Naissance

0.50

 cavalli

0.47

0.45

<eos>

0.43

⋱

0.42

verifyException

0.42

########.

0.41

Activations Density 0.044%

mathematical constructs and notation

No Comments

No Known Activations