INDEX

Explanations

code-like sequences or handles with special characters

oai_token-act-pair · gpt-3.5-turbo

special characters or symbols, particularly those representing a specific sentiment or tone

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.2.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Manit

-0.72

 Whitman

-0.71

wagen

-0.71

 Akin

-0.68

 gloom

-0.67

 Gillespie

-0.66

 recording

-0.65

Pug

-0.65

 slip

-0.65

 motion

-0.64

POSITIVE LOGITS

ðŁ

1.32

¥ŀ

1.24

į

1.19

Ĵ

1.16

ĵ

1.12

Į

1.09

ĵĺ

1.08

Ķ

1.07

ĻĤ

1.06

ĩ

1.05

Activations Density 0.004%

code-like sequences or handles with special characters

special characters or symbols, particularly those representing a specific sentiment or tone

No Comments

No Known Activations