INDEX

Explanations

technical descriptions and references to diagrams or figures in documents

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

thon

-0.15

ystone

-0.14

ScreenState

-0.14

ianne

-0.13

 Virgin

-0.13

jour

-0.13

anus

-0.13

pery

-0.13

 Enum

-0.13

 Least

-0.13

POSITIVE LOGITS

 impression

0.16

0.15

oir

0.15

.Timer

0.14

ramids

0.14

Ð¼Ð¾Ð²

0.14

HEMA

0.14

ħ§

0.14

 ramen

0.14

oplay

0.13

Activations Density 0.023%

technical descriptions and references to diagrams or figures in documents

No Comments

No Known Activations