INDEX

Explanations

phrases related to technical capabilities or performance

oai_token-act-pair · gpt-3.5-turbo

terms related to scalability and usability

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

ulla

-0.78

inas

-0.75

sonian

-0.74

ergy

-0.72

eric

-0.71

inis

-0.69

lements

-0.69

jon

-0.68

yton

-0.66

stad

-0.66

POSITIVE LOGITS

 mismatch

1.02

anship

0.91

ability

0.90

 assurance

0.88

iencies

0.84

worthiness

0.84

 parity

0.84

iqueness

0.82

orously

0.82

ulner

0.79

Activations Density 0.075%

phrases related to technical capabilities or performance

terms related to scalability and usability

No Comments

No Known Activations