INDEX

Explanations

locations or settings within written text

oai_token-act-pair · gpt-3.5-turbo

the word "inside" and related context about locations or situations

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

umbn

-0.65

 dear

-0.63

 convey

-0.62

venge

-0.60

 lord

-0.59

WR

-0.59

Ye

-0.58

 professionally

-0.58

 colony

-0.58

 favour

-0.57

POSITIVE LOGITS

neath

0.87

itialized

0.82

xtap

0.80

Iterator

0.78

ctr

0.72

statement

0.72

views

0.70

aucus

0.69

lash

0.68

asionally

0.67

Activations Density 0.065%

locations or settings within written text

the word "inside" and related context about locations or situations

No Comments

No Known Activations