INDEX

Explanations

punctuation marks and quotation marks in texts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.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

ãĥ¼ãĥĨãĤ£

-0.15

PLUS

-0.15

inand

-0.14

iaux

-0.14

hti

-0.14

plusplus

-0.14

 objective

-0.14

ernaut

-0.14

akter

-0.14

 Plus

-0.14

POSITIVE LOGITS

 another

0.19

 Another

0.18

Another

0.17

 perhaps

0.16

 beyond

0.16

 Perhaps

0.16

Of

0.16

 Beyond

0.15

Perhaps

0.15

 parallel

0.15

Activations Density 0.189%

punctuation marks and quotation marks in texts

No Comments

No Known Activations