INDEX

Explanations

references to the duration of time in seasons, particularly emphasizing seasons that have a strong association

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "seasons."

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

ymes

-0.69

 procedural

-0.69

umbn

-0.67

ens

-0.67

ilde

-0.67

apa

-0.66

elight

-0.64

ople

-0.61

rote

-0.60

ech

-0.60

POSITIVE LOGITS

ago

1.03

hips

0.91

poons

0.83

Ago

0.81

 spanning

0.81

pring

0.78

 thereafter

0.77

 consecut

0.72

 totaling

0.72

ettings

0.69

Activations Density 0.057%

references to the duration of time in seasons, particularly emphasizing seasons that have a strong association

occurrences of the word "seasons."

No Comments

No Known Activations