INDEX

Explanations

instances of conjunctions or phrases that indicate connections or relationships

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 17-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_17/width_16k/average_l0_77

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.17.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

 westfalen

-0.59

cijas

-0.58

neh

-0.56

 robi

-0.55

 يتيمه

-0.54

 infuse

-0.54

catching

-0.53

inske

-0.53

JsonHelper

-0.52

wrote

-0.52

POSITIVE LOGITS

 taken

0.87

 flown

0.86

 eaten

0.82

 shown

0.80

 placed

0.77

 thrown

0.76

 seen

0.75

 kept

0.74

 presented

0.72

 given

0.70

Activations Density 0.482%

instances of conjunctions or phrases that indicate connections or relationships

No Comments

No Known Activations