INDEX

Explanations

phrases indicating exclusivity or limitation

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 18-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L18R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.18.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

anguage

-0.15

oka

-0.15

ãĥ¼ãĤ¯

-0.15

alez

-0.15

simp

-0.15

oshi

-0.15

 Bills

-0.15

olley

-0.14

isty

-0.14

Ð¾Ð·Ð°

-0.14

POSITIVE LOGITS

few

0.23

 handful

0.21

few

0.19

one

0.19

Few

0.18

Few

0.18

 mÃ¡lo

0.16

ecta

0.15

Canonical

0.14

 HOLDER

0.14

Activations Density 0.095%

phrases indicating exclusivity or limitation

No Comments

No Known Activations