INDEX

Explanations

conjunctions and connecting words in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 11-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L11R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.11.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

wit

-0.16

 monster

-0.15

mun

-0.14

lÃ©

-0.14

Verdana

-0.13

adder

-0.13

 autom

-0.13

edn

-0.13

uti

-0.13

apur

-0.13

POSITIVE LOGITS

kker

0.16

ovsky

0.15

idan

0.14

_AB

0.14

antha

0.14

zan

0.13

åľ³

0.13

yl

0.13

uhl

0.13

gate

0.13

Activations Density 0.080%

conjunctions and connecting words in various contexts

No Comments

No Known Activations