INDEX

Explanations

occurrences of the abbreviation "e.g." which indicates examples or elaboration 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 @ 11-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L11R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.11.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

ady

-0.15

Ð²ÑĸÐ»ÑĮ

-0.15

ar

-0.15

.yahoo

-0.14

lar

-0.14

aye

-0.14

åĭ¢

-0.13

nel

-0.13

ToMany

-0.13

ray

-0.13

POSITIVE LOGITS

ained

0.15

-alist

0.15

otr

0.14

oti

0.14

iad

0.14

(TypeError

0.14

hari

0.14

inton

0.14

Ã¤Ã¤

0.14

-i

0.14

Activations Density 0.016%

occurrences of the abbreviation "e.g." which indicates examples or elaboration in text

No Comments

No Known Activations