INDEX

Explanations

short phrases indicating end or conclusion in a context

oai_token-act-pair · gpt-3.5-turbo

phrases that emphasize the concept of "all" and completeness

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ngth

-0.78

çīĪ

-0.72

kefeller

-0.67

alks

-0.66

rams

-0.66

moil

-0.65

clock

-0.63

nor

-0.63

apons

-0.62

prus

-0.61

POSITIVE LOGITS

SPONSORED

0.81

PLIED

0.74

 coincidence

0.73

soType

0.70

natureconservancy

0.70

 explan

0.67

IER

0.67

 blasphemy

0.65

tru

0.65

 scary

0.64

Activations Density 0.280%

short phrases indicating end or conclusion in a context

phrases that emphasize the concept of "all" and completeness

No Comments

No Known Activations