INDEX

Explanations

phrases describing limitations and drawbacks 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 @ 4-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_4/width_16k/average_l0_124

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.4.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

ächter

-0.48

aid

-0.47

 urmă

-0.45

 stră

-0.43

зиру

-0.43

 mukaan

-0.42

ştır

-0.42

 Heber

-0.41

ówczas

-0.41

 veramente

-0.41

POSITIVE LOGITS

SourceChecksum

0.76

 lack

0.72

VersionUID

0.72

 inability

0.69

qrstuvwxyz

0.68

 nature

0.68

BufferException

0.67

 estimés

0.66

 availability

0.65

jenost

0.65

Activations Density 0.952%

phrases describing limitations and drawbacks in various contexts

No Comments

No Known Activations