INDEX

Explanations

references to statements made by spokespersons or officials

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 9-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L9R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.9.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

 Norris

-0.15

slash

-0.14

 wrappers

-0.14

 Ð¸Ñģ

-0.14

è¦ª

-0.13

à¹ĩà¸ĩ

-0.13

sug

-0.13

erals

-0.13

-N

-0.13

oras

-0.13

POSITIVE LOGITS

 Ramp

0.16

tak

0.15

ISTR

0.15

inear

0.15

esimal

0.15

nid

0.15

ROY

0.14

.Invariant

0.14

ubat

0.14

Whe

0.14

Activations Density 0.018%

references to statements made by spokespersons or officials

No Comments

No Known Activations