INDEX

Explanations

words related to technology products and companies

oai_token-act-pair · gpt-3.5-turbo

occurrences of specific prefixes and suffixes in words

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

BALL

-0.74

 allowances

-0.73

angan

-0.72

icho

-0.70

ģĸ

-0.67

Weather

-0.67

ADRA

-0.66

ccording

-0.66

 srfAttach

-0.66

ometimes

-0.66

POSITIVE LOGITS

theless

0.97

phrine

0.70

lus

0.70

opian

0.69

termin

0.68

ukong

0.66

anasia

0.65

 grav

0.64

phant

0.63

oder

0.62

Activations Density 0.091%

words related to technology products and companies

occurrences of specific prefixes and suffixes in words

No Comments

No Known Activations