INDEX

Explanations

phrases indicating uncertainty or speculation, often starting with "Who knows"

oai_token-act-pair · gpt-3.5-turbo

expressions of uncertainty or curiosity

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

ItemTracker

-0.77

ciating

-0.77

phrine

-0.76

herent

-0.66

inance

-0.66

etsk

-0.64

REM

-0.62

ructose

-0.61

packages

-0.61

charges

-0.61

POSITIVE LOGITS

how

0.70

 darn

0.70

fri

0.67

how

0.67

 whats

0.65

 scen

0.64

 srfAttach

0.63

 geop

0.61

lege

0.61

ãĤ½

0.60

Activations Density 0.030%

phrases indicating uncertainty or speculation, often starting with "Who knows"

expressions of uncertainty or curiosity

No Comments

No Known Activations