INDEX

Explanations

medical conditions and technical terms related to equipment

oai_token-act-pair · gpt-3.5-turbo

specific keywords related to low levels or declining trends 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

SPONSORED

-0.69

frey

-0.68

ADRA

-0.66

ophob

-0.66

MODE

-0.65

 Gate

-0.63

 Gleaming

-0.62

Pal

-0.61

ove

-0.61

çīĪ

-0.60

POSITIVE LOGITS

 incidence

0.82

(<

0.77

acco

0.73

effic

0.72

eco

0.71

jri

0.69

abor

0.68

 wast

0.68

 pesky

0.66

 concentration

0.64

Activations Density 0.279%

medical conditions and technical terms related to equipment

specific keywords related to low levels or declining trends in various contexts

No Comments

No Known Activations