INDEX

Explanations

specific mentions of locations or positions, such as geographical places or spatial references

oai_token-act-pair · gpt-3.5-turbo

references to specific locations or places

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

vous

-0.84

jin

-0.80

missions

-0.77

tails

-0.76

ework

-0.76

sonian

-0.76

unes

-0.75

xtap

-0.73

sle

-0.69

rahim

-0.69

POSITIVE LOGITS

 location

0.97

 locations

0.96

 locate

0.77

 locality

0.76

 nearest

0.76

URA

0.73

 scout

0.72

 vicinity

0.70

 landmarks

0.69

 where

0.68

Activations Density 0.068%

specific mentions of locations or positions, such as geographical places or spatial references

references to specific locations or places

No Comments

No Known Activations