INDEX

Explanations

character names and their interactions in a dialogue-heavy context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 15-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L15R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.15.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

 blinked

-0.17

 blink

-0.16

blink

-0.15

 blinking

-0.15

sx

-0.15

Î³ÎºÎ¿

-0.15

pitch

-0.14

elpers

-0.14

ello

-0.14

mate

-0.14

POSITIVE LOGITS

ideshow

0.15

ktor

0.14

akhir

0.14

getDescription

0.14

anmar

0.14

NESS

0.13

bette

0.13

.CompareTo

0.13

ivers

0.13

sto

0.13

Activations Density 0.314%

character names and their interactions in a dialogue-heavy context

No Comments

No Known Activations