INDEX

Explanations

references to types and classifications

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

sizeCache

-0.91

AddTagHelper

-0.88

 nicio

-0.87

 strconv

-0.86

Erreferentziak

-0.84

 McCullough

-0.83

UnsafeEnabled

-0.82

 DialogInterface

-0.78

 ',',

-0.78

ISR

-0.78

POSITIVE LOGITS

 Type

2.12

 type

2.10

 TYPE

2.07

 types

1.88

type

1.83

TYPE

1.82

Type

1.76

 tipo

1.67

types

1.62

 Types

1.59

Activations Density 0.166%

references to types and classifications

No Comments

No Known Activations