INDEX

Explanations

punctuation and commas used within contexts related to lists and explanations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 15-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L15R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.15.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

ampus

-0.16

iej

-0.15

switch

-0.14

lear

-0.14

akan

-0.14

-ie

-0.14

rouw

-0.14

agna

-0.14

èº«ä¸Ĭ

-0.13

 Humanities

-0.13

POSITIVE LOGITS

its

0.19

ReadStream

0.18

0.16

upal

0.16

hol

0.16

Î¹Î±ÏĤ

0.15

ksen

0.15

 which

0.14

 both

0.14

 First

0.14

Activations Density 0.154%

punctuation and commas used within contexts related to lists and explanations

No Comments

No Known Activations