INDEX

Explanations

references to authors, editors, and contributors in academic contexts

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ãĥ¼ãĥ«

-0.17

_ck

-0.15

AGO

-0.15

 ç§ĭ

-0.15

-fat

-0.15

fat

-0.14

Ø¹ÙĬ

-0.14

Ð¾Ð²Ðµ

-0.14

akash

-0.14

Fat

-0.14

POSITIVE LOGITS

Sharper

0.15

uke

0.15

ski

0.14

hl

0.14

avel

0.14

mac

0.14

 sinking

0.14

elementType

0.13

.createClass

0.13

emas

0.13

Activations Density 0.285%

references to authors, editors, and contributors in academic contexts

No Comments

No Known Activations