INDEX

Explanations

words related to specific locations, possibly within a community or setting

oai_token-act-pair · gpt-3.5-turbo

key nouns related to geography, governance, and institutions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

enegger

-0.66

\\\\\\\\

-0.59

selves

-0.56

Ĭ±

-0.56

 Instr

-0.54

itored

-0.53

Tid

-0.53

Redd

-0.52

ECA

-0.52

ecause

-0.52

POSITIVE LOGITS

 fallacy

0.74

iest

0.72

ultimate

0.64

 theorem

0.62

 embodiment

0.62

 hypothesis

0.61

 planner

0.61

 keyword

0.61

 skyline

0.60

kernel

0.60

Activations Density 0.597%

words related to specific locations, possibly within a community or setting

key nouns related to geography, governance, and institutions

No Comments

No Known Activations