INDEX

Explanations

computer code and technical information

oai_token-act-pair · gpt-3.5-turbo

symbols and formatting indicators related to data or documentation

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

 footing

-0.74

 simplicity

-0.67

 connectivity

-0.66

TPPStreamerBot

-0.63

ileaks

-0.63

 piping

-0.62

 inev

-0.61

 outsider

-0.61

odon

-0.61

 braking

-0.61

POSITIVE LOGITS

sbm

0.99

channelAvailability

0.87

ãĥł

0.78

 Edited

0.75

 Reserved

0.74

MSN

0.71

 Converted

0.70

ategory

0.67

Rated

0.66

ãĥĻ

0.64

Activations Density 0.244%

computer code and technical information

symbols and formatting indicators related to data or documentation

No Comments

No Known Activations