INDEX

Explanations

references to oral activities

oai_token-act-pair · gpt-3.5-turbo

references to oral activities or experiences

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

 guiActiveUnfocused

-0.76

IELD

-0.73

âĸĪâĸĪâĸĪâĸĪâĸĪâĸĪâĸĪâĸĪ

-0.72

âĻ¥

-0.71

Clar

-0.69

Charlie

-0.68

 Shining

-0.68

Ross

-0.68

AMI

-0.67

âĸ¬

-0.67

POSITIVE LOGITS

 oral

1.40

 cavity

0.98

 orally

0.98

 Oral

0.97

 ingestion

0.96

 contraceptive

0.92

 contraceptives

0.88

 hygiene

0.86

ibaba

0.85

muc

0.83

Activations Density 0.007%

references to oral activities

references to oral activities or experiences

No Comments

No Known Activations