INDEX

Explanations

elements related to publishing and documentation

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 16-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L16R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.16.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

-prepend

-0.15

 Coch

-0.14

izarre

-0.14

#echo

-0.14

_prep

-0.13

aman

-0.13

aves

-0.13

Od

-0.13

ìłľ

-0.13

çģµ

-0.13

POSITIVE LOGITS

ading

0.16

Ð¾ÑĤÑĮ

0.16

ç±į

0.15

iguous

0.14

ibur

0.14

eding

0.14

nts

0.13

onio

0.13

onse

0.13

Activations Density 0.127%

elements related to publishing and documentation

No Comments

No Known Activations