INDEX

Explanations

descriptive adjectives that convey quality or attention-grabbing characteristics

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

strup

-0.15

Ð±Ñĭ

-0.14

rikes

-0.14

esa

-0.14

ponsor

-0.14

ias

-0.14

 chamber

-0.14

/gif

-0.14

arah

-0.13

ħn

-0.13

POSITIVE LOGITS

åŃĺäºİ

0.15

adder

0.15

ragen

0.14

ycz

0.14

emente

0.14

ÙĤØ¯Ø±

0.14

icus

0.14

âĢĮØªØ±ÛĮÙĨ

0.14

ialis

0.14

iban

0.14

Activations Density 0.286%

descriptive adjectives that convey quality or attention-grabbing characteristics

No Comments

No Known Activations