INDEX

Explanations

key terms related to logic and reasoning in arguments

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 13-gemmascope-res-65k

Configuration

google/gemma-scope-2b-pt-res/layer_13/width_65k/average_l0_75

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

Data Type

float32

Hook Name

blocks.13.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

ThroughAttribute

-0.89

 كومونز

-0.86

 ExecuteAsync

-0.85

twimg

-0.83

 AssemblyCompany

-0.82

 архивлан

-0.81

__(/*!

-0.79

writeField

-0.77

IsContent

-0.77

TestingModule

-0.77

POSITIVE LOGITS

to

0.79

 reasons

0.63

 ways

0.61

Reasons

0.59

why

0.56

 Reasons

0.56

 Gründe

0.55

for

0.54

 moyen

0.53

 Ways

0.53

Activations Density 0.162%

key terms related to logic and reasoning in arguments

No Comments

No Known Activations