INDEX

Explanations

proper nouns and character names, particularly in narrative contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 25-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L25R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.25.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

icans

-0.18

plain

-0.17

uality

-0.16

stit

-0.15

bilt

-0.15

ysis

-0.14

urs

-0.14

ãģªãģĦ

-0.14

urge

-0.14

holders

-0.14

POSITIVE LOGITS

-Ãªtre

0.17

xious

0.16

/do

0.16

 Snowden

0.15

-ahead

0.14

ìĤ¬íķŃ

0.14

naments

0.14

ropic

0.14

itia

0.14

Activations Density 0.262%

proper nouns and character names, particularly in narrative contexts

No Comments

No Known Activations