INDEX

Explanations

terms related to final outcomes and goals in various contexts

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

 BrowserModule

-0.53

 antaranya

-0.52

 nakalista

-0.50

getragen

-0.50

écart

-0.49

 JpaRepository

-0.49

layın

-0.48

sedown

-0.47

åp

-0.46

 puissent

-0.46

POSITIVE LOGITS

 destination

1.20

 destinations

1.04

 outcome

0.98

destination

0.96

 goal

0.93

 outcomes

0.92

 Destination

0.90

Destination

0.90

 destino

0.89

 Outcome

0.87

Activations Density 0.724%

terms related to final outcomes and goals in various contexts

No Comments

No Known Activations