INDEX

Explanations

explicit mentions of the word "remember"

oai_token-act-pair · gpt-3.5-turbo

instances of the word "remember" and related variations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

amen

-0.80

\\\\\\\\\\\\\\\\

-0.71

oit

-0.70

quart

-0.68

onica

-0.68

iliary

-0.67

otomy

-0.67

assi

-0.65

erb

-0.64

ECD

-0.64

POSITIVE LOGITS

remember

1.07

 remembered

0.93

 remembering

0.92

 remembrance

0.89

 vividly

0.83

abilia

0.81

theless

0.80

 remember

0.80

 memories

0.80

 recollection

0.80

Activations Density 0.014%

explicit mentions of the word "remember"

instances of the word "remember" and related variations

No Comments

No Known Activations