INDEX

Explanations

specific terms related to emotions or internal states

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 29-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L29R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.29.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

 Ð¿Ð¾Ð±Ð°

-0.18

 fore

-0.18

 Ð´Ð¾ÐºÑĥÐ¼

-0.17

 Ð¸Ð½ÑĦÐ¾ÑĢÐ¼Ð°

-0.17

cr

-0.17

 mach

-0.17

le

-0.17

 Ð³ÑĢÐ°Ð´Ñĥ

-0.16

ga

-0.16

POSITIVE LOGITS

 addCriterion

0.22

 Ð²

0.22

 Ð¸

0.21

Ð²

0.18

$MESS

0.17

âĦĸâĦĸ

0.17

 Ð¶Ðµ

0.17

 Ð½Ðµ

0.16

.scalablytyped

0.16

undry

0.16

Activations Density 0.361%

specific terms related to emotions or internal states

No Comments

No Known Activations