INDEX

Explanations

instances of 'posted' along with surrounding context

oai_token-act-pair · gpt-3.5-turbo

instances of the word "posted" and its variations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

osi

-0.75

ppo

-0.72

 Flavoring

-0.69

schild

-0.65

 Bauer

-0.64

apy

-0.64

 Oaks

-0.62

 Brill

-0.62

kos

-0.62

rica

-0.62

POSITIVE LOGITS

ulate

0.98

hum

0.93

 online

0.92

mortem

0.88

 anonymously

0.86

 flyers

0.84

doctoral

0.84

 postings

0.81

onymous

0.77

URL

0.76

Activations Density 0.049%

instances of 'posted' along with surrounding context

instances of the word "posted" and its variations

No Comments

No Known Activations