INDEX

Explanations

terms related to "cavalry" or "cavalry-related activities

oai_token-act-pair · gpt-3.5-turbo

references to cavalry and related terms or events

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

bered

-0.73

 Myster

-0.73

skirts

-0.70

hov

-0.68

gement

-0.67

zee

-0.66

ging

-0.64

jah

-0.63

nov

-0.63

prime

-0.63

POSITIVE LOGITS

 Caval

1.06

cade

0.92

aval

0.91

isson

0.79

iere

0.76

 chorus

0.74

assis

0.71

riott

0.70

illon

0.70

 buffs

0.69

Activations Density 0.020%

terms related to "cavalry" or "cavalry-related activities

references to cavalry and related terms or events

No Comments

No Known Activations