INDEX

Explanations

names of universities or colleges

oai_token-act-pair · gpt-3.5-turbo

proper nouns, particularly names of people and locations

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

raints

-0.76

Ø¯

-0.74

aneously

-0.69

 squats

-0.67

"-

-0.67

ively

-0.64

ÙĦ

-0.64

heimer

-0.63

rikes

-0.62

DERR

-0.61

POSITIVE LOGITS

stal

1.06

Bry

0.99

sta

0.91

llo

0.87

utral

0.86

cipled

0.84

giene

0.82

gments

0.81

sonian

0.81

cle

0.81

Activations Density 0.014%

names of universities or colleges

proper nouns, particularly names of people and locations

No Comments

No Known Activations