INDEX

Explanations

names of medical professionals

oai_token-act-pair · gpt-3.5-turbo

occurrences of names followed by a period

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

 capped

-0.71

 gearing

-0.69

thumbnails

-0.65

 capitals

-0.65

 unpop

-0.65

 suspending

-0.64

 redundancy

-0.64

 servicing

-0.64

 piping

-0.64

 dissolution

-0.63

POSITIVE LOGITS

 Jill

1.14

 Canaver

1.12

 Robert

1.05

 Kenneth

1.02

 Stephen

1.01

 Jeffrey

1.01

 Lawrence

1.00

 Cornel

1.00

 Phillip

1.00

 Andrew

0.99

Activations Density 0.039%

names of medical professionals

occurrences of names followed by a period

No Comments

No Known Activations