INDEX

Explanations

adjectives and descriptions that denote quality, size, and type

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 23-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_23/width_16k/average_l0_75

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.23.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

uitable

-0.74

 חיצוניים

-0.71

 InputDecoration

-0.68

abetes

-0.66

 مرئيه

-0.65

CompilerServices

-0.63

principalColumn

-0.62

WebElementEntity

-0.61

ruptedException

-0.61

yntaxException

-0.60

POSITIVE LOGITS

 priced

1.02

 sized

1.00

 looking

0.99

 sounding

0.96

 LOOKING

0.84

 appearing

0.81

 shaped

0.80

 smelling

0.78

looking

0.77

 faced

0.76

Activations Density 0.450%

adjectives and descriptions that denote quality, size, and type

No Comments

No Known Activations