INDEX

Explanations

words related to locations, specifically names of places

oai_token-act-pair · gpt-3.5-turbo

references to the letter 'k' in various contexts

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

 mosqu

-0.70

 liberties

-0.65

 celebr

-0.64

 Nero

-0.62

 unlaw

-0.61

 emergencies

-0.59

 Murdoch

-0.59

 cryst

-0.58

 dividend

-0.58

 clich

-0.57

POSITIVE LOGITS

ansas

1.34

orea

1.14

uyomi

1.14

irk

1.13

ileaks

1.12

idding

1.10

rieg

1.09

ngth

1.03

icker

1.02

regate

1.02

Activations Density 0.062%

words related to locations, specifically names of places

references to the letter 'k' in various contexts

No Comments

No Known Activations