INDEX

Explanations

unique identifiers or symbols in a technical or scientific context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 31-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L31R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.31.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

adio

-0.20

monds

-0.16

ASM

-0.15

igue

-0.15

 accounted

-0.15

kÃ¡

-0.15

igu

-0.14

FileInfo

-0.14

 accountId

-0.14

asma

-0.14

POSITIVE LOGITS

hurst

0.20

mem

0.17

sk

0.16

 invited

0.16

mem

0.15

Mem

0.15

 enjoying

0.15

 already

0.15

 enjoyment

0.15

 membranes

0.15

Activations Density 0.007%

unique identifiers or symbols in a technical or scientific context

No Comments

No Known Activations