INDEX

Explanations

words related to explosions or strong impacts

oai_token-act-pair · gpt-3.5-turbo

instances and references to explosive events or blasts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

guyen

-0.78

 compr

-0.70

ADRA

-0.68

éĸ

-0.67

sit

-0.67

Lean

-0.67

Govern

-0.67

BOOK

-0.65

sie

-0.65

PN

-0.64

POSITIVE LOGITS

 furnace

1.09

ographed

0.91

efully

0.89

asts

0.87

ocalypse

0.85

 furn

0.85

asted

0.84

 blasts

0.82

anche

0.79

 blast

0.79

Activations Density 0.035%

words related to explosions or strong impacts

instances and references to explosive events or blasts

No Comments

No Known Activations