INDEX

Explanations

instances of locations or positional references

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 23-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_23/width_16k/average_l0_75

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.23.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

 befind

-0.57

]='\

-0.56

 terms

-0.52

="#"><

-0.50

 Between

-0.48

 steder

-0.48

 NSCoder

-0.48

 жить

-0.48

standig

-0.48

 TERMS

-0.47

POSITIVE LOGITS

 response

0.70

 court

0.68

 retaliation

0.68

 كومونز

0.66

 reaction

0.58

 protest

0.57

 comments

0.57

 connection

0.56

 apparent

0.54

 favor

0.53

Activations Density 0.354%

instances of locations or positional references

No Comments

No Known Activations