INDEX

Explanations

numerical identifiers and references in various contexts, particularly related to legislative bills and districts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 7-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L7R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.7.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

udeau

-0.15

osomal

-0.15

lez

-0.15

å¶

-0.14

Ø³ØªÙħ

-0.14

lds

-0.14

uckle

-0.14

RP

-0.14

icolon

-0.14

ucursal

-0.14

POSITIVE LOGITS

ÑģÑĮÐ¾Ð³Ð¾

0.15

ElementType

0.15

ous

0.14

Means

0.14

xit

0.13

hus

0.13

flu

0.13

itters

0.13

 Fritz

0.13

 Means

0.13

Activations Density 0.233%

numerical identifiers and references in various contexts, particularly related to legislative bills and districts

No Comments

No Known Activations