INDEX

Explanations

phrases related to characters or individuals in stories or narratives

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

.createServer

-0.16

hores

-0.16

 Unidos

-0.15

à¤¿à¤¹

-0.15

Responder

-0.15

leur

-0.15

okers

-0.14

ëª°

-0.14

rength

-0.14

uhl

-0.14

POSITIVE LOGITS

 whose

0.18

fur

0.16

 everybody

0.16

ekk

0.16

 everyone

0.16

¯

0.16

whose

0.15

endon

0.15

yre

0.14

Activations Density 0.189%

phrases related to characters or individuals in stories or narratives

No Comments

No Known Activations