INDEX

Explanations

references to archived content or archives

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 7-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L7R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.7.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

InSection

-0.16

avec

-0.16

 breed

-0.15

.gridView

-0.15

amel

-0.15

 tiÃªn

-0.14

osh

-0.14

 McDon

-0.14

wend

-0.13

stal

-0.13

POSITIVE LOGITS

ripper

0.18

itect

0.17

stroy

0.15

ekim

0.15

ulated

0.15

celed

0.15

AGMA

0.15

iciel

0.15

interop

0.14

tvrt

0.14

Activations Density 0.018%

references to archived content or archives

No Comments

No Known Activations