INDEX

Explanations

punctuation and transitions in writing

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 22-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L22R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.22.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

lland

-0.16

ALER

-0.15

me

-0.15

 halves

-0.15

ANCELED

-0.14

 together

-0.14

jerne

-0.14

AccessException

-0.13

bjerg

-0.13

eor

-0.13

POSITIVE LOGITS

 readers

0.16

Last

0.16

ucc

0.16

 Earlier

0.16

esterday

0.16

>Last

0.16

 reader

0.15

 Last

0.15

 sometimes

0.15

remember

0.15

Activations Density 0.007%

punctuation and transitions in writing

No Comments

No Known Activations