INDEX

Explanations

Instances where something is being reminded to someone

oai_token-act-pair · gpt-3.5-turbo

references to personal memories or associations

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

advertising

-0.68

<?

-0.67

soever

-0.65

 disagrees

-0.64

mouth

-0.63

aird

-0.63

hement

-0.62

econom

-0.62

chen

-0.60

atts

-0.59

POSITIVE LOGITS

 vividly

0.83

 nostalg

0.77

 Invaders

0.67

 srfAttach

0.67

 compr

0.66

 heroism

0.66

 chant

0.66

 recall

0.66

 Recall

0.64

 remembering

0.64

Activations Density 0.085%

Instances where something is being reminded to someone

references to personal memories or associations

No Comments

No Known Activations