INDEX

Explanations

phrases and structures related to explanations and reasons

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 15-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L15R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.15.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

aira

-0.15

	Copyright

-0.15

.providers

-0.14

ewan

-0.13

Ð¸Ñģ

-0.13

ees

-0.13

edia

-0.13

bab

-0.12

POSITIVE LOGITS

âĳł

0.15

yat

0.15

.First

0.15

agus

0.15

:↵

0.15

chie

0.14

:↵↵↵↵

0.14

ãģ²ãģ¨

0.13

tg

0.13

:↵↵

0.13

Activations Density 0.089%

phrases and structures related to explanations and reasons

No Comments

No Known Activations