INDEX

Explanations

occurrences of the word "for."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 0-gemmascope-mlp-65k

Configuration

google/gemma-scope-2b-pt-mlp/layer_0/width_65k/average_l0_72

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

Data Type

float32

Hook Name

blocks.0.hook_mlp_out

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

InjectAttribute

-1.05

#+#

-1.05

tagHelperRunner

-0.98

 Numerade

-0.96

 Roskov

-0.94

 שוליים

-0.93

principalTable

-0.90

+#+

-0.90

 kasarigan

-0.90

expandindo

-0.89

POSITIVE LOGITS

 poches

0.49

0.48

0.47

 vettor

0.46

0.45

 fantasma

0.45

ack

0.45

0.44

 buone

0.43

Activations Density 0.270%

occurrences of the word "for."

No Comments

No Known Activations