INDEX

Explanations

phrases related to locations or places

oai_token-act-pair · gpt-3.5-turbo

instances of critical or significant information, particularly about programs or events

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

 patience

-0.87

 stability

-0.82

 retire

-0.80

 endurance

-0.78

enary

-0.77

 stabilization

-0.76

 regener

-0.74

 heal

-0.74

 sustainability

-0.74

 regeneration

-0.73

POSITIVE LOGITS

 According

1.05

According

1.02

Refer

1.00

 excerpts

0.97

 purported

0.96

 purportedly

0.93

Sources

0.91

 Specifically

0.90

 Sources

0.90

SPONSORED

0.90

Activations Density 0.758%

phrases related to locations or places

instances of critical or significant information, particularly about programs or events

No Comments

No Known Activations