INDEX

Explanations

words related to geographical locations or entities

oai_token-act-pair · gpt-3.5-turbo

words or terms related to specific locations or entities, particularly geographic or cultural references

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

-0.76

 Mechdragon

-0.69

%]

-0.66

estyles

-0.59

NK

-0.58

 partName

-0.58

orders

-0.57

çīĪ

-0.56

 Princ

-0.56

arter

-0.56

POSITIVE LOGITS

pillar

0.88

pill

0.78

ccording

0.77

isure

0.77

hurst

0.77

hetics

0.76

haar

0.74

collar

0.73

mentioned

0.72

ILA

0.72

Activations Density 0.073%

words related to geographical locations or entities

words or terms related to specific locations or entities, particularly geographic or cultural references

No Comments

No Known Activations