INDEX

Explanations

qualifiers that express smallness or a lack of intensity

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

 ModelExpression

-0.89

ReusableCell

-0.84

ⓧ

-0.83

 kasarigan

-0.74

rrggbb

-0.73

 Fiske

-0.72

WriteTagHelper

-0.71

 MainAxisSize

-0.70

TypedDataSet

-0.69

 MacLeod

-0.69

POSITIVE LOGITS

bit

1.05

 bisschen

0.97

 biraz

0.85

 کمی

0.81

 trochę

0.81

Slightly

0.73

 Kome

0.72

 dụ

0.71

 slightly

0.71

Bit

0.71

Activations Density 0.079%

qualifiers that express smallness or a lack of intensity

No Comments

No Known Activations