INDEX

Explanations

inquiries and prompts related to user input or programming questions

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

reib

-0.16

kees

-0.14

oto

-0.14

cul

-0.14

abeth

-0.14

 intermittent

-0.13

abi

-0.13

oom

-0.13

_ib

-0.13

ntax

-0.13

POSITIVE LOGITS

croft

0.17

andes

0.15

onica

0.15

ailable

0.15

sdale

0.15

aley

0.14

@student

0.14

 chÃŃ

0.14

oldem

0.14

PRESS

0.14

Activations Density 0.004%

inquiries and prompts related to user input or programming questions

No Comments

No Known Activations