INDEX

Explanations

references to geographical and cultural contexts involving surveillance or monitoring

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 dostan

-0.15

 prostÅĻednictvÃŃm

-0.14

aliz

-0.14

 blÃŃ

-0.12

alars

-0.12

ÏĢÎ¬

-0.12

IALIZ

-0.12

Î»Î¯

-0.12

ÙĳØ¯

-0.12

 tedy

-0.11

POSITIVE LOGITS

by

1.21

 oleh

0.93

 ØªÙĪØ³Ø·

0.82

 bá»Łi

0.77

by

0.73

_by

0.68

 tarafÄ±ndan

0.55

.by

0.54

 ìĿĺíķ´

0.52

çĶ±

0.51

Activations Density 1.927%

references to geographical and cultural contexts involving surveillance or monitoring

No Comments

No Known Activations