INDEX

Explanations

the conjunction "or"

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L1R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.1.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

lec

-0.18

 ones

-0.16

Ã²ng

-0.15

ÑĥÑĢÐ³

-0.15

irit

-0.14

lector

-0.14

ency

-0.14

oder

-0.13

tlement

-0.13

POSITIVE LOGITS

phans

0.17

usi

0.16

nown

0.16

ally

0.15

thon

0.14

 sost

0.14

medi

0.14

naments

0.14

ebin

0.14

mdir

0.13

Activations Density 0.221%

the conjunction "or"

No Comments

No Known Activations