INDEX

Explanations

references to telling stories or narratives 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 @ 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

intptr

-0.14

itz

-0.14

ilibrium

-0.13

ety

-0.13

 Titles

-0.13

ska

-0.13

ÑĩÐ°ÑĤ

-0.13

 stabilize

-0.13

_INF

-0.13

.spotify

-0.12

POSITIVE LOGITS

 story

0.93

 Story

0.79

story

0.76

 stories

0.76

 STORY

0.73

Story

0.72

-story

0.69

_story

0.66

 tale

0.63

.story

0.63

Activations Density 0.328%

references to telling stories or narratives in various contexts

No Comments

No Known Activations