INDEX

Explanations

instances where someone's actions lead to unexpected or ironic outcomes

oai_token-act-pair · gpt-3.5-turbo

the word "only" 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 @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

insula

-0.77

ahime

-0.76

hement

-0.66

idon

-0.66

iosyncr

-0.59

endo

-0.59

lass

-0.59

sem

-0.59

intensity

-0.57

 Code

-0.56

POSITIVE LOGITS

 marginally

0.94

 kidding

0.93

 incidentally

0.89

 seconds

0.77

WARE

0.77

ĨĴ

0.72

 lasts

0.69

Owner

0.68

ices

0.67

 minutes

0.67

Activations Density 0.062%

instances where someone's actions lead to unexpected or ironic outcomes

the word "only" in various contexts

No Comments

No Known Activations