INDEX

Explanations

numerical designations formatted as 'a number' followed by 'v'

oai_token-act-pair · gpt-3.5-turbo

terms related to various forms of identity and classification

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

laden

-0.73

 guid

-0.73

 Obst

-0.69

yards

-0.69

cair

-0.69

REDACTED

-0.68

 sidx

-0.68

mington

-0.67

 Ridley

-0.64

wind

-0.64

POSITIVE LOGITS

clusive

0.69

 affair

0.69

 snowball

0.67

 Movement

0.66

 auth

0.64

etimes

0.63

cation

0.62

ĥ

0.61

imensional

0.61

 Cinema

0.61

Activations Density 0.242%

numerical designations formatted as 'a number' followed by 'v'

terms related to various forms of identity and classification

No Comments

No Known Activations