INDEX

Explanations

phrases indicating a sense of logic or rationale

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 4-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L4R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.4.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

unks

-0.15

esian

-0.15

 Rover

-0.15

Haz

-0.14

ervas

-0.14

odi

-0.14

ostream

-0.14

iÄħ

-0.13

elt

-0.13

plets

-0.13

POSITIVE LOGITS

BILE

0.17

orde

0.16

ibold

0.16

unte

0.16

è¼ª

0.15

ToDevice

0.15

ÑĢÑĥÐº

0.15

ToFit

0.15

UpDown

0.14

conc

0.14

Activations Density 0.243%

phrases indicating a sense of logic or rationale

No Comments

No Known Activations

phrases indicating a sense of logic or rationale

No Comments

No Known Activations