INDEX

Explanations

the word "cross"

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "cross" in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

Dur

-0.71

Peb

-0.70

 Policies

-0.68

TION

-0.68

lessness

-0.67

onom

-0.66

idity

-0.64

Lav

-0.64

Reason

-0.64

payers

-0.64

POSITIVE LOGITS

 cross

3.76

 Cross

2.59

Cross

2.56

cross

2.43

 crosses

2.10

 crossed

1.87

 crossing

1.62

 crossover

1.27

 crossings

1.23

 Crossing

1.06

Activations Density 0.009%

the word "cross"

occurrences of the word "cross" in various contexts

No Comments

No Known Activations