INDEX

Explanations

phrases emphasizing the complete absence or negation of something

oai_token-act-pair · gpt-3.5-turbo

phrases emphasizing the absence or irrelevance of something

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

seless

-0.88

ku

-0.83

ahime

-0.77

pat

-0.75

using

-0.75

irs

-0.74

spect

-0.73

fil

-0.71

used

-0.71

ÅĤ

-0.71

POSITIVE LOGITS

 whatsoever

1.03

soever

1.03

 else

0.88

 except

0.85

THING

0.85

 resembling

0.84

 ensued

0.83

 transpired

0.82

 existed

0.79

 happened

0.77

Activations Density 0.016%

phrases emphasizing the complete absence or negation of something

phrases emphasizing the absence or irrelevance of something

No Comments

No Known Activations