INDEX

Explanations

expressions that indicate consistency in statements or themes across 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 @ 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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

kuuta

-0.80

 Graff

-0.73

PMailer

-0.72

 McCullough

-0.72

 estimés

-0.71

MigrationBuilder

-0.71

reactstrap

-0.70

 VIDEOTAPE

-0.67

jsonwebtoken

-0.67

IALES

-0.67

POSITIVE LOGITS

 consistent

1.91

 Consistency

1.87

consistent

1.86

 consistency

1.82

Consistency

1.81

 Consistent

1.78

Consistent

1.72

consistency

1.70

 inconsistent

1.56

 inconsistency

1.50

Activations Density 0.118%

expressions that indicate consistency in statements or themes across various contexts

No Comments

No Known Activations