INDEX

Explanations

phrases related to philosophical or moral discussions

oai_token-act-pair · gpt-3.5-turbo

statements concerning the nature and definition of concepts, particularly relating to religion and morality

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

 throats

-0.81

çīĪ

-0.73

itches

-0.72

 Horizons

-0.71

sers

-0.70

igham

-0.69

Brow

-0.68

ilst

-0.67

 Monitor

-0.65

bilt

-0.64

POSITIVE LOGITS

 synonymous

1.14

 shorthand

1.08

 insepar

0.99

 intrinsically

0.98

 meaningless

0.96

 subjective

0.95

 inherently

0.89

 derived

0.86

 rooted

0.86

 transformative

0.86

Activations Density 0.348%

phrases related to philosophical or moral discussions

statements concerning the nature and definition of concepts, particularly relating to religion and morality

No Comments

No Known Activations