INDEX

Explanations

keywords indicating cause and effect relationships in various contexts

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 12-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L12R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.12.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

IFrame

Link

Negative Logits

Ñĳ

-0.14

elsen

-0.14

cri

-0.14

NotSupportedException

-0.13

 RuntimeObject

-0.13

road

-0.13

Portrait

-0.13

.organ

-0.12

-basic

-0.12

POSITIVE LOGITS

zb

0.16

ewise

0.15

0.14

wise

0.14

obus

0.14

onso

0.14

aphael

0.13

voj

0.13

odus

0.13

iggs

0.13

Activations Density 0.287%

keywords indicating cause and effect relationships in various contexts

No Known Activations