INDEX

Explanations

mentions of appendices and supplementary materials in documents

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 12-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L12R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.12.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

MODULE

-0.15

rokes

-0.15

gap

-0.14

hiba

-0.14

 GenerationType

-0.14

yd

-0.14

 Amazon

-0.13

jak

-0.13

/bit

-0.13

.Quad

-0.13

POSITIVE LOGITS

irst

0.17

icon

0.16

iams

0.15

rang

0.15

ieg

0.14

prav

0.14

STRU

0.14

umlu

0.14

es

0.14

imes

0.14

Activations Density 0.024%

mentions of appendices and supplementary materials in documents

No Comments

No Known Activations