INDEX

Explanations

terms related to different wavelengths of light

oai_token-act-pair · gpt-3.5-turbo

terms related to different types of light and their properties

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

wright

-0.76

ndra

-0.74

agall

-0.69

bare

-0.69

steen

-0.68

jury

-0.68

hoff

-0.68

piece

-0.67

makers

-0.67

gaard

-0.66

POSITIVE LOGITS

 wavelengths

1.10

 radiation

1.07

rared

1.06

 ultraviolet

1.04

 rays

0.95

velength

0.95

raviolet

0.89

 infrared

0.88

iolet

0.88

orescence

0.83

Activations Density 0.026%

terms related to different wavelengths of light

terms related to different types of light and their properties

No Comments

No Known Activations