INDEX

Explanations

words related to geopolitical locations, specifically Tehran

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "Tehran."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

 peer

-0.63

environment

-0.61

ause

-0.61

 Hispanic

-0.60

otrop

-0.60

dom

-0.59

hu

-0.59

ipp

-0.59

 intangible

-0.58

ONSORED

-0.58

POSITIVE LOGITS

 Tehran

2.66

 Ankara

2.64

 Islamabad

2.46

 Riyadh

2.39

 Kiev

2.36

 Kabul

2.06

 Moscow

2.06

 Pyongyang

2.01

 Baghdad

1.99

 Cairo

1.98

Activations Density 0.032%

words related to geopolitical locations, specifically Tehran

occurrences of the word "Tehran."

No Comments

No Known Activations