INDEX

Explanations

instances of the letter 'A' in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 16-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L16R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.16.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

vsp

-0.16

Ð¼Ñĭ

-0.16

loo

-0.15

Ñī

-0.14

igid

-0.14

oulouse

-0.13

èª¿

-0.13

éģĭåĭķ

-0.13

ward

-0.13

Suffix

-0.13

POSITIVE LOGITS

anje

0.17

tainment

0.16

ajs

0.14

canf

0.14

izyon

0.14

 Cape

0.14

ê³Ħ

0.14

\Context

0.14

ohl

0.14

izada

0.13

Activations Density 0.003%

instances of the letter 'A' in various contexts

No Comments

No Known Activations