INDEX

Explanations

phrases that indicate relationships between concepts or conditions and their implications

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 6-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L6R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.6.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

amba

-0.19

opies

-0.14

Ð½Ð¾Ñĩ

-0.14

azo

-0.13

auer

-0.13

Ī

-0.13

_SAFE

-0.13

hk

-0.13

è¡

-0.13

ubes

-0.13

POSITIVE LOGITS

stal

0.15

ering

0.15

ï¸ı

0.15

Ð¾Ð»ÑĮÐ½Ð¾

0.14

strup

0.14

 recess

0.14

 companion

0.14

0.13

uÅŁ

0.13

Activations Density 0.617%

phrases that indicate relationships between concepts or conditions and their implications

No Comments

No Known Activations