INDEX

Explanations

words related to locations or geographical regions

oai_token-act-pair · gpt-3.5-turbo

references to energy-related concepts

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Archdemon

-0.77

itures

-0.72

arium

-0.65

ipeg

-0.63

urdue

-0.62

 declass

-0.62

ufact

-0.62

umerable

-0.61

ittees

-0.60

éĹĺ

-0.60

POSITIVE LOGITS

gie

1.01

rics

0.96

cock

0.93

roxy

0.85

gy

0.85

psy

0.85

gian

0.84

gets

0.84

gins

0.81

rex

0.80

Activations Density 0.011%

words related to locations or geographical regions

references to energy-related concepts

No Comments

No Known Activations