INDEX

Explanations

references to physical positioning or orientation

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 23-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L23R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.23.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

Äĥm

-0.15

ãģ¨ãģĵãĤį

-0.15

yster

-0.15

-0.14

zan

-0.14

mediate

-0.14

vik

-0.14

uta

-0.14

_macros

-0.14

POSITIVE LOGITS

 front

0.83

front

0.60

 Front

0.60

 rear

0.59

 FRONT

0.58

Front

0.57

-front

0.57

 fronts

0.54

_front

0.53

 frontend

0.53

Activations Density 0.103%

references to physical positioning or orientation

No Comments

No Known Activations

references to physical positioning or orientation

No Comments

No Known Activations