INDEX

Explanations

medical terms related to a specific condition, likely "oid"

oai_token-act-pair · gpt-3.5-turbo

terms related to various types of "oids" or "steroids."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

 Chap

-0.71

 continuing

-0.65

Chr

-0.61

 Kens

-0.61

 Cert

-0.59

 pars

-0.59

 sentencing

-0.58

hon

-0.57

Kyr

-0.57

 opportunities

-0.57

POSITIVE LOGITS

oid

4.59

oids

4.32

oidal

2.83

ormal

1.14

omorph

1.11

ylon

1.09

osaurus

1.04

aceae

1.01

otype

1.01

otropic

1.01

Activations Density 0.014%

medical terms related to a specific condition, likely "oid"

terms related to various types of "oids" or "steroids."

No Comments

No Known Activations