INDEX

Explanations

clusters of similar events or activities

oai_token-act-pair · gpt-3.5-turbo

terms indicating multiple occurrences or quantities of something

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

 Dominion

-0.64

estial

-0.63

 Rebels

-0.61

olves

-0.60

rador

-0.60

 Sector

-0.58

 Federation

-0.57

 habitable

-0.57

atable

-0.56

 colonies

-0.56

POSITIVE LOGITS

ãĥ¥

0.86

Timeout

0.85

ãĤ¤ãĥĪ

0.84

urst

0.81

SHARE

0.79

otom

0.79

 Scroll

0.74

kok

0.73

stakes

0.73

ãĤ´ãĥ³

0.73

Activations Density 0.016%

clusters of similar events or activities

terms indicating multiple occurrences or quantities of something

No Comments

No Known Activations