INDEX

Explanations

text indicating a prompt to verify that the reader is not a robot

oai_token-act-pair · gpt-3.5-turbo

references to the word "you" in various contexts

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

 Canaver

-0.91

 Kang

-0.67

 Aberdeen

-0.66

 Wonderland

-0.66

 Pratt

-0.65

 Cater

-0.64

achable

-0.63

Lau

-0.62

 Defenders

-0.62

acular

-0.62

POSITIVE LOGITS

're

1.33

've

1.13

'll

1.06

RS

1.02

hei

0.92

 guys

0.87

 know

0.85

'd

0.83

can

0.83

tub

0.82

Activations Density 0.136%

text indicating a prompt to verify that the reader is not a robot

references to the word "you" in various contexts

No Comments

No Known Activations