INDEX

Explanations

instances of the word "space" and its variations in relation to context or constraints

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 7-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_7/width_16k/average_l0_69

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.7.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

 Monfieur

-1.01

 zelve

-0.99

(!__

-0.94

 hydrauli

-0.94

__":

-0.93

 aveug

-0.93

 themſelves

-0.93

UnusedPrivate

-0.92

quelles

-0.88

 ainfi

-0.88

POSITIVE LOGITS

 space

2.06

 spaces

1.92

 Space

1.91

 SPACE

1.82

Space

1.77

space

1.75

 Spaces

1.74

Spaces

1.72

SPACE

1.66

spaces

1.63

Activations Density 0.040%

instances of the word "space" and its variations in relation to context or constraints

No Comments

No Known Activations