INDEX

Explanations

occurrences of the word "feedback" and related concepts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 0-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_0/width_16k/average_l0_105

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.0.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

 Shakspeare

-1.00

 auffi

-0.88

 Shaksp

-0.84

 raiſ

-0.83

 gynhyrchwyd

-0.83

 leaſt

-0.82

 CHAPITRE

-0.82

 myſelf

-0.82

 resourceCulture

-0.82

 Theſe

-0.80

POSITIVE LOGITS

Sal

0.69

 opinion

0.63

 color

0.57

 feedback

0.55

dropdown

0.52

熟

0.51

ban

0.51

Sal

0.49

ਿ

0.48

 Opinion

0.48

Activations Density 0.121%

occurrences of the word "feedback" and related concepts

No Comments

No Known Activations