INDEX

Explanations

phrases or words related to physical climbing or ascension

oai_token-act-pair · gpt-3.5-turbo

the act of climbing, both literally and metaphorically

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

Hart

-0.80

ãĥĨãĤ£

-0.76

rums

-0.73

tering

-0.70

pora

-0.69

rum

-0.69

invoke

-0.66

esson

-0.64

OH

-0.64

pling

-0.64

POSITIVE LOGITS

 Everest

1.04

 stairs

1.01

 footh

0.93

 climb

0.93

 climbing

0.92

 climbers

0.90

 climbs

0.84

 ladder

0.84

 aboard

0.83

 ascent

0.81

Activations Density 0.033%

phrases or words related to physical climbing or ascension

the act of climbing, both literally and metaphorically

No Comments

No Known Activations