INDEX

Explanations

terms related to reading, information sharing, and legal processes

oai_token-act-pair · gpt-3.5-turbo

references to reading and the concept of company

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

ycle

-0.74

jriwal

-0.67

cele

-0.66

userc

-0.66

 suicide

-0.62

 multif

-0.61

kefeller

-0.61

 shortcut

-0.61

anos

-0.61

 progression

-0.58

POSITIVE LOGITS

 Borders

0.74

 Pearce

0.66

ographers

0.64

pants

0.61

Ã±a

0.60

forts

0.59

ography

0.57

 Angelo

0.57

NPR

0.56

 Sanct

0.56

Activations Density 0.390%

terms related to reading, information sharing, and legal processes

references to reading and the concept of company

No Comments

No Known Activations