INDEX

Explanations

conjunctions and transitional phrases that indicate reasoning or cause-and-effect relationships

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 4-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L4R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.4.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

lops

-0.16

arus

-0.15

rott

-0.15

sis

-0.15

anke

-0.15

tery

-0.14

orsk

-0.14

ught

-0.14

raries

-0.14

POSITIVE LOGITS

forth

0.30

ìį¨

0.20

acer

0.17

ful

0.16

adio

0.16

acro

0.15

acers

0.14

ester

0.14

ç¤

0.14

hatt

0.14

Activations Density 0.008%

conjunctions and transitional phrases that indicate reasoning or cause-and-effect relationships

No Comments

No Known Activations