INDEX

Explanations

dates written in a specific format (month, day, year) combined with specific usernames

oai_token-act-pair · gpt-3.5-turbo

commas in the text

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

 expansions

-0.85

 detectors

-0.76

 predec

-0.72

 successors

-0.68

 expansion

-0.67

 connections

-0.66

 unnecess

-0.66

avorite

-0.65

 glim

-0.64

 superiors

-0.64

POSITIVE LOGITS

0.95

0.90

0.89

0.88

0.85

0.84

0.82

0.81

Activations Density 0.051%

dates written in a specific format (month, day, year) combined with specific usernames

commas in the text

No Comments

No Known Activations