INDEX

Explanations

medical-related terms or actions, particularly related to medications or treatments

oai_token-act-pair · gpt-3.5-turbo

words related to dispensing and the effects of substances

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

ULTS

-0.72

cius

-0.66

Pwr

-0.65

CHR

-0.65

 Nare

-0.60

---------

-0.59

EAR

-0.59

 ACTION

-0.58

erno

-0.58

 Rabbit

-0.58

POSITIVE LOGITS

ing

2.55

eering

1.31

ingham

1.25

ership

1.17

edIn

1.16

ING

1.15

ingly

1.12

ation

1.11

ed

1.08

ging

1.07

Activations Density 0.151%

medical-related terms or actions, particularly related to medications or treatments

words related to dispensing and the effects of substances

No Comments

No Known Activations