INDEX

Explanations

adjectives describing surprise or unexpectedness

oai_token-act-pair · gpt-3.5-turbo

instances of the word "surprisingly"

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

Negative Logits

tein

-0.86

abad

-0.83

rie

-0.75

flight

-0.73

icipated

-0.71

allas

-0.71

gang

-0.71

iche

-0.71

amins

-0.69

ingham

-0.69

POSITIVE LOGITS

 surprisingly

0.87

ESA

0.79

 surprising

0.75

 Jagu

0.74

Buyable

0.72

 enough

0.72

NESS

0.70

 surpr

0.69

 srfAttach

0.69

 absent

0.69

Activations Density 0.010%

adjectives describing surprise or unexpectedness

instances of the word "surprisingly"

No Known Activations