INDEX

Explanations

references or terminology related to academic research and scholarly work

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 4-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L4R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.4.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

onse

-0.16

KY

-0.15

(strtolower

-0.14

enze

-0.14

azen

-0.14

terra

-0.14

 recess

-0.14

entai

-0.13

 paternal

-0.13

POSITIVE LOGITS

Cue

0.15

 sÃ¡ch

0.15

gart

0.15

asto

0.14

ensburg

0.14

cheid

0.14

udget

0.14

-ie

0.14

_thumb

0.14

ãģĮãģĦ

0.14

Activations Density 0.442%

references or terminology related to academic research and scholarly work

No Comments

No Known Activations