INDEX

Explanations

expressions related to improvement and accessibility

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

yntaxException

-0.74

MessageOf

-0.72

 Theſe

-0.64

 GenerationType

-0.64

 {}),

-0.60

BufferException

-0.58

 []),

-0.58

 propOrder

-0.57

Према

-0.57

__))

-0.57

POSITIVE LOGITS

 AssemblyCulture

0.73

fjspx

0.70

WriteTagHelper

0.61

verwijspagina

0.56

RTEE

0.53

quelize

0.49

__.__

0.47

 becomes

0.46

Ổ

0.46

PyExc

0.45

Activations Density 0.194%

expressions related to improvement and accessibility

No Comments

No Known Activations