INDEX

Explanations

phrases related to completeness or inclusion

oai_token-act-pair · gpt-3.5-turbo

phrases emphasizing relationships and connections

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ug

-0.74

(?,

-0.72

lander

-0.70

rio

-0.70

different

-0.63

DEN

-0.62

raft

-0.62

itiveness

-0.61

hate

-0.61

earable

-0.61

POSITIVE LOGITS

 exceptions

0.65

 whichever

0.64

 evidenced

0.61

cir

0.61

 perhaps

0.61

 Emin

0.59

İĭ

0.58

 ascending

0.57

 consultation

0.56

ä»£

0.56

Activations Density 0.399%

phrases related to completeness or inclusion

phrases emphasizing relationships and connections

No Comments

No Known Activations