INDEX

Explanations

instances of conjunctions and related phrases that join ideas or entities

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 16-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L16R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.16.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

IFrame

Link

Negative Logits

-scalable

-0.16

vÃ¡l

-0.15

cher

-0.15

deen

-0.15

urai

-0.14

Î¼Î¿Î¯

-0.14

opot

-0.14

Ð¾Ð¿Ð¾Ð»

-0.14

_IW

-0.14

okie

-0.14

POSITIVE LOGITS

/from

0.20

/of

0.16

/or

0.15

 either

0.14

å¾ħ

0.14

 EITHER

0.14

ernel

0.13

ÙĲÙĬ

0.13

paren

0.13

resp

0.13

Activations Density 0.084%

instances of conjunctions and related phrases that join ideas or entities

No Known Activations