INDEX

Explanations

numerical quantities or emphasized descriptions

oai_token-act-pair · gpt-3.5-turbo

the repeated use of the word "one" to indicate a singular concept or entity

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

osponsors

-0.71

hips

-0.71

inders

-0.69

ãĤµ

-0.68

ãĤŃ

-0.68

ooks

-0.67

ãĤ½

-0.67

gif

-0.66

inas

-0.64

older

-0.64

POSITIVE LOGITS

 hundred

0.95

 sided

0.86

 Hundred

0.85

 thing

0.84

 dimensional

0.83

rency

0.80

 embodiment

0.75

 wonders

0.75

 thousand

0.74

 Piece

0.70

Activations Density 0.127%

numerical quantities or emphasized descriptions

the repeated use of the word "one" to indicate a singular concept or entity

No Comments

No Known Activations