INDEX

Explanations

references to reading, viewing, or reproducing text and related content

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 21-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L21R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.21.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

á»£

-0.13

Ã¦r

-0.13

Å¡ti

-0.12

geme

-0.12

eshire

-0.12

hari

-0.12

hq

-0.12

htar

-0.12

alama

-0.12

Ã©nÃ©

-0.11

POSITIVE LOGITS

 full

0.94

 complete

0.84

 entire

0.83

å®Įæķ´

0.77

full

0.75

 whole

0.73

-full

0.73

 FULL

0.73

 entirety

0.69

 completo

0.69

Activations Density 0.587%

references to reading, viewing, or reproducing text and related content

No Comments

No Known Activations