INDEX

Explanations

words related to actions or policies that influence accessibility, trust, or productivity in various contexts

oai_token-act-pair · gpt-3.5-turbo

words related to trust and access to resources or services

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

iera

-0.65

eger

-0.64

kos

-0.63

arthed

-0.60

mma

-0.57

ernel

-0.57

mega

-0.56

SPONSORED

-0.56

oscope

-0.55

adies

-0.55

POSITIVE LOGITS

ahime

0.71

itism

0.68

 altogether

0.66

ivity

0.65

 Takeru

0.65

 morale

0.64

 gains

0.63

.–

0.62

 eligibility

0.62

 barriers

0.61

Activations Density 0.340%

words related to actions or policies that influence accessibility, trust, or productivity in various contexts

words related to trust and access to resources or services

No Comments

No Known Activations