INDEX

Explanations

words related to military ranks and titles, particularly "Sgt"

oai_token-act-pair · gpt-3.5-turbo

repeated references to the title "Sgt" followed by a name or number

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ahime

-0.88

rights

-0.77

theless

-0.75

GPU

-0.71

worm

-0.69

PUT

-0.69

stage

-0.69

FORE

-0.68

hur

-0.67

uador

-0.65

POSITIVE LOGITS

geant

1.27

Sgt

1.06

 Sergeant

0.97

gt

0.86

 Pepper

0.84

 sergeant

0.81

utenant

0.80

chief

0.75

veland

0.75

 Duty

0.75

Activations Density 0.018%

words related to military ranks and titles, particularly "Sgt"

repeated references to the title "Sgt" followed by a name or number

No Comments

No Known Activations