INDEX

Explanations

punctuation and structural markers in text

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 21-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L21R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.21.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

inux

-0.16

egl

-0.15

æ³°

-0.14

.fast

-0.14

ington

-0.14

çĶ°

-0.14

ÙİØ§ÙĨ

-0.14

_UNIX

-0.14

.wr

-0.13

eters

-0.13

POSITIVE LOGITS

akin

0.17

 suff

0.16

bsites

0.15

gh

0.14

à¤ĸ

0.14

 works

0.14

 Maiden

0.14

 Works

0.13

aden

0.13

Activations Density 0.017%

punctuation and structural markers in text

No Comments

No Known Activations