INDEX

Explanations

code-related structures and syntax, particularly function definitions and error handling

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

enterOuterAlt

-0.55

ochem

-0.48

<%=

-0.44

nji

-0.44

ინ

-0.44

nó

-0.41

ceros

-0.41

esos

-0.41

شون

-0.41

obar

-0.41

POSITIVE LOGITS

 self

3.13

self

3.04

 Self

2.43

Self

2.29

 SELF

2.16

 selves

2.00

 herself

1.92

SELF

1.88

 Selbst

1.85

 zelf

1.83

Activations Density 0.053%

code-related structures and syntax, particularly function definitions and error handling

No Comments

No Known Activations