INDEX

Explanations

verbs related to being located or situated somewhere

oai_token-act-pair · gpt-3.5-turbo

phrases that indicate location or existence 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

aud

-0.74

aldi

-0.68

panic

-0.68

obs

-0.68

smart

-0.67

iles

-0.67

agi

-0.67

 household

-0.66

ilar

-0.65

iling

-0.64

POSITIVE LOGITS

uten

0.80

lie

0.78

rences

0.75

 squarely

0.75

utenant

0.74

 dormant

0.73

 behind

0.70

wic

0.70

behind

0.70

 Reincarn

0.69

Activations Density 0.016%

verbs related to being located or situated somewhere

phrases that indicate location or existence of something

No Comments

No Known Activations