INDEX

Explanations

phrases related to discussions or talk about experiences and changes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 11-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L11R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.11.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

rung

-0.15

street

-0.15

oho

-0.14

ucci

-0.14

 part

-0.14

oh

-0.14

roids

-0.14

roman

-0.14

á»¯

-0.13

 Vanguard

-0.13

POSITIVE LOGITS

unar

0.17

 indeed

0.16

akan

0.15

chia

0.14

eam

0.14

 crossorigin

0.14

_JOIN

0.14

à¹Ģà¸¥à¸¢

0.14

LATED

0.14

olation

0.14

Activations Density 0.253%

phrases related to discussions or talk about experiences and changes

No Comments

No Known Activations