INDEX

Explanations

phrases related to how something is being treated or perceived

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "treat" in various contexts related to actions or behaviors

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

 Frie

-0.67

ItemTracker

-0.65

Bi

-0.60

 Falcons

-0.59

Rae

-0.59

 Vide

-0.58

 Starg

-0.58

 recount

-0.57

Mo

-0.57

 Bucc

-0.57

POSITIVE LOGITS

ises

1.10

terson

0.93

ttes

0.91

ments

0.90

onom

0.86

ise

0.85

imeters

0.83

ts

0.81

tons

0.80

ties

0.80

Activations Density 0.018%

phrases related to how something is being treated or perceived

occurrences of the word "treat" in various contexts related to actions or behaviors

No Comments

No Known Activations