INDEX

Explanations

measurements and tools related to assessing various aspects or qualities

oai_token-act-pair · gpt-3.5-turbo

terms related to measurement and evaluation

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

 Fathers

-0.85

uality

-0.76

scape

-0.73

heid

-0.71

sis

-0.71

orig

-0.68

Lay

-0.68

————————

-0.67

 Origins

-0.67

edia

-0.66

POSITIVE LOGITS

gau

1.38

 gauge

1.27

 buoy

0.86

 Pens

0.80

 stations

0.80

Gau

0.76

UGE

0.74

 pigeon

0.71

chell

0.71

 station

0.71

Activations Density 0.007%

measurements and tools related to assessing various aspects or qualities

terms related to measurement and evaluation

No Comments

No Known Activations