INDEX

Explanations

instances of the word "follow" and its variations in different 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

Yag

-0.80

 Pemberton

-0.75

oldValue

-0.74

 Urqu

-0.74

mbolos

-0.71

 Eisenberg

-0.69

 McCarty

-0.68

ñón

-0.66

 AppModule

-0.66

 Jacobsen

-0.66

POSITIVE LOGITS

follow

2.10

 follow

2.09

 Follows

2.04

Follow

2.04

 follows

2.01

 Follow

2.00

 FOLLOW

1.97

follows

1.85

FOLLOW

1.74

 Followed

1.71

Activations Density 0.123%

instances of the word "follow" and its variations in different contexts

No Comments

No Known Activations