INDEX

Explanations

structured references to leadership concepts or discussions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.24.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

©

-0.14

iese

-0.14

ble

-0.14

raw

-0.14

 shading

-0.14

ographed

-0.14

aret

-0.14

CAP

-0.13

MG

-0.13

decrypt

-0.13

POSITIVE LOGITS

 schema

0.20

schema

0.19

VIA

0.18

-schema

0.18

 Neutral

0.16

UBY

0.16

EntryPoint

0.16

/schema

0.16

onto

0.16

Neutral

0.15

Activations Density 0.004%

structured references to leadership concepts or discussions

No Comments

No Known Activations