INDEX

Explanations

capitalized acronyms

oai_token-act-pair · gpt-3.5-turbo

the abbreviation "OM" used 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 @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

chnology

-0.81

ivities

-0.78

wall

-0.77

hs

-0.74

issance

-0.71

connected

-0.68

draw

-0.67

ive

-0.66

ience

-0.65

wright

-0.63

POSITIVE LOGITS

EGA

1.02

edia

0.91

orrow

0.90

obile

0.85

atis

0.79

ento

0.78

ESA

0.78

ETH

0.78

essage

0.78

ilitary

0.76

Activations Density 0.025%

capitalized acronyms

the abbreviation "OM" used in various contexts

No Comments

No Known Activations