INDEX

Explanations

phrases related to actions and their consequences, particularly in a causal context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.20.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 linkovi

-0.75

 AssemblyCulture

-0.74

ThroughAttribute

-0.72

 estekak

-0.68

 useAppContext

-0.68

Tikang

-0.66

сылкі

-0.65

Geplaatst

-0.64

UnusedPrivate

-0.63

 كومونز

-0.63

POSITIVE LOGITS

 justice

0.92

justice

0.74

 Justice

0.70

Justice

0.70

 JUSTICE

0.69

 service

0.61

 harm

0.59

service

0.58

 wonders

0.57

 justicia

0.56

Activations Density 0.140%

phrases related to actions and their consequences, particularly in a causal context

No Comments

No Known Activations