INDEX

Explanations

negations or denials paired with adjectives

oai_token-act-pair · gpt-3.5-turbo

phrases emphasizing absence or lack of something

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

orks

-0.80

flies

-0.74

yx

-0.68

fs

-0.68

UME

-0.67

rib

-0.66

olds

-0.66

haul

-0.66

die

-0.64

 Chains

-0.64

POSITIVE LOGITS

 else

0.79

 hidden

0.77

 buried

0.72

 intrinsic

0.70

 objectionable

0.70

 overlap

0.70

 happening

0.69

 poetic

0.68

 lurking

0.68

 shameful

0.67

Activations Density 0.044%

negations or denials paired with adjectives

phrases emphasizing absence or lack of something

No Comments

No Known Activations