INDEX

Explanations

key concepts related to various categories of classifications and entities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 15-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L15R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.15.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

oui

-0.16

ertoire

-0.16

isha

-0.14

sembling

-0.14

ÏĦÎ·Î½

-0.14

ÑĢÐµÐ¶Ð´

-0.13

igators

-0.13

 Astr

-0.13

Ã´

-0.13

 Jenkins

-0.13

POSITIVE LOGITS

ucz

0.17

 Mounted

0.15

 laid

0.14

/on

0.14

ibs

0.14

ikit

0.14

isse

0.13

 incom

0.13

sie

0.13

/mobile

0.13

Activations Density 0.020%

key concepts related to various categories of classifications and entities

No Comments

No Known Activations