INDEX

Explanations

the word "building"

oai_token-act-pair · gpt-3.5-turbo

mentions of buildings, specifically "building 9" and related references

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

IFrame

Link

Negative Logits

romeda

-0.94

opard

-0.82

ï¸ı

-0.79

mia

-0.78

nir

-0.76

 Nadu

-0.74

ternity

-0.74

cipline

-0.73

ican

-0.73

sha

-0.72

POSITIVE LOGITS

raper

1.04

 blocks

0.97

 renovations

0.91

 occupancy

0.88

building

0.87

 constructed

0.86

 buildings

0.86

 erected

0.84

 skysc

0.79

 Buildings

0.79

Activations Density 0.031%

the word "building"

mentions of buildings, specifically "building 9" and related references

No Known Activations