INDEX

Explanations

terms related to editing and writing processes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 6-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L6R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.6.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

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

-0.15

rag

-0.15

rup

-0.15

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

-0.14

ogs

-0.14

echa

-0.14

auga

-0.14

nder

-0.14

oggles

-0.14

letcher

-0.13

POSITIVE LOGITS

Ð¾Ð¶

0.16

Lit

0.15

antino

0.15

ihan

0.15

icine

0.14

heading

0.14

iba

0.14

FUN

0.13

nila

0.13

Activations Density 0.317%

terms related to editing and writing processes

No Comments

No Known Activations