INDEX

Explanations

names of people, particularly in relation to their relationships and social dynamics

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

smr

-0.16

 Coastal

-0.16

asics

-0.15

egin

-0.15

_lite

-0.14

UpperCase

-0.14

okie

-0.14

elts

-0.13

 hereby

-0.13

.intent

-0.13

POSITIVE LOGITS

 caption

0.19

 dating

0.19

 datings

0.17

isay

0.16

 statuses

0.16

 Kardashian

0.16

 captions

0.16

 romant

0.16

Caption

0.16

 dated

0.15

Activations Density 0.112%

names of people, particularly in relation to their relationships and social dynamics

No Comments

No Known Activations