INDEX

Explanations

code-related constructs, particularly those associated with string and dictionary manipulations in a programming context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 5-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_5/width_16k/average_l0_68

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.5.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

providedIn

-0.71

-0.61

sidemargin

-0.61

-0.60

TemporalType

-0.59

ever

-0.59

ej

-0.59

-0.57

—

-0.57

POSITIVE LOGITS

NSString

2.52

 NSString

2.11

 myſelf

1.25

Efq

1.20

 itſelf

1.18

 themſelves

1.16

 himſelf

1.12

 iſt

1.09

 QString

1.09

 ſeveral

1.06

Activations Density 0.035%

code-related constructs, particularly those associated with string and dictionary manipulations in a programming context

No Comments

No Known Activations